Correspondence
Anesthesia Awareness and the Bispectral Index
N Engl J Med 2008; 359:427-431July 24, 2008
- Article
To the Editor:
Avidan et al. (March 13 issue)1 confirm that bispectral index (BIS)–guided care achieves a 0.2% incidence of awareness during anesthesia in high-risk patients, as previously reported,2 but further conclusions and recommendations are not supported. Despite the investigators' original hypothesis “that the incidence of awareness [with an end-tidal protocol] will be equivalent to or lower” than that with a BIS-guided protocol,3 the sample-size calculation erroneously assumed no treatment effect for the anesthetic protocol (Table 1Table 1
Appropriate Sample Sizes for the Comparison of Two Interventions to Avoid Intraoperative Awareness.). Consequently, this study had an 80% probability of missing a 50% difference in the relative efficacy of the two interventions.4 Unexplained gaps in BIS trends suggest that inadequate training or vigilance and poor protocol compliance contributed to both of the awareness cases in the BIS-guided group and impeded reduced anesthesia dosing that is typically seen with BIS monitoring.5 Furthermore, missing perioperative-management and postoperative-outcome data in this vulnerable population limits generalization.
In the accompanying editorial,6 Orser should have highlighted the limitations of an underpowered study4 and considered the avoidance of awareness within the broader applications of brain monitoring. A single inadequate study testing an alternative intervention does not justify sweeping conclusions that question the value of BIS monitoring and that are neither supported by data nor consistent with the preponderance of clinical evidence.5
Scott D. Kelley, M.D.
Paul J. Manberg, Ph.D.
Jeff C. Sigl, Ph.D.
Aspect Medical Systems, Norwood, MA 02062
skelley@aspectms.com Drs. Kelley, Manberg, and Sigl report being employees of Aspect Medical Systems, which manufactures BIS monitoring systems, and holding equity interest and stock options in the company. No other potential conflict of interest relevant to this letter was reported.
6 References1
Avidan MS ,Zhang L ,Burnside BA , et al. Anesthesia awareness and the bispectral index. N Engl J Med 2008;358:1097-1108
Full Text | Web of Science | Medline2
Myles PS ,Leslie K ,McNeil J ,Forbes A ,Chan MT . Bispectral index monitoring to prevent awareness during anaesthesia: the B-Aware randomised controlled trial. Lancet 2004;363:1757-1763
CrossRef | Web of Science | Medline3
Avidan M, Burnside B. Study evaluating ways of preventing patients from being awake during high-risk surgery and anesthesia. St. Louis: Washington University School of Medicine, Barnes-Jewish Foundation, 2008. (Accessed July 7, 2008, at http://clinicaltrials.gov/ct2/show/NCT00281489.)
4
Freiman JA ,Chalmers TC ,Smith H Jr ,Kuebler RR . The importance of beta, the type II error and sample size in the design and interpretation of the randomized control trial: survey of 71 “negative” trials. N Engl J Med 1978;299:690-694
Full Text | Web of Science | Medline5
Punjasawadwong Y ,Boonjeungmonkol N ,Phongchiewboon A . Bispectral index for improving anaesthetic delivery and postoperative recovery. Cochrane Database Syst Rev 2007;4:CD003843-CD003843
Medline6
Orser BA . Depth-of-anesthesia monitor and the frequency of intraoperative awareness. N Engl J Med 2008;358:1189-1191
Full Text | Web of Science | Medline
To the Editor:
Avidan et al. could not demonstrate a reduction in the incidence of awareness with BIS monitoring. This should come as no surprise, since they enrolled patients who had a lower risk of awareness than those in our study1 (e.g., unselected patients undergoing open-heart surgery). Also, they included patients with “minor” risk factors. Such liberal inclusion criteria will lead to a reduction in trial events, as did occur (four events). This severely limits the capacity of the study to provide reliable results. This is best appreciated when considering the wide 95% confidence interval that Avidan et al. used for the difference between groups (−0.56% to 0.57%). An exact 95% confidence interval for the odds ratio ranges from 1/14 to 14, and since awareness is rare, this is an approximate 95% confidence interval for the relative risk reduction. So a 14-fold benefit in either direction cannot be ruled out. This study cannot provide reliable information on which to base anesthetic practice.2 An updated meta-analysis including other trials1,3 suggests a likely benefit of BIS monitoring (odds ratio, 0.31; 95% confidence interval, 0.11 to 0.91; P=0.03) (Figure 1Figure 1
Updated Meta-Analysis of Bispectral Index (BIS) versus Traditional Anesthetic Monitoring.).Paul S. Myles, M.P.H., M.D.
Alfred Hospital, Melbourne, VIC 3004, Australia
p.myles@alfred. org. au Kate Leslie, M.D., M.Epi.
Royal Melbourne Hospital, Melbourne, VIC 3000, AustraliaAndrew Forbes, M.Sc., Ph.D.
Monash University, Melbourne, VIC 3004, AustraliaDrs. Myles, Leslie, and Forbes report receiving loan equipment and unrestricted funding from Aspect Medical Systems. No other potential conflict of interest relevant to this letter was reported.
3 References1
Myles PS ,Leslie K ,McNeil J ,Forbes A ,Chan MT . Bispectral index monitoring to prevent awareness during anaesthesia: the B-Aware randomised controlled trial. Lancet 2004;363:1757-1763
CrossRef | Web of Science | Medline2
Tunis SR ,Stryer DB ,Clancy CM . Practical clinical trials: increasing the value of clinical research for decision making in clinical and health policy. JAMA 2003;290:1624-1632
CrossRef | Web of Science | Medline3
Puri GD ,Murthy SS . Bispectral index monitoring in patients undergoing cardiac surgery under cardiopulmonary bypass. Eur J Anaesthesiol 2003;20:451-456
CrossRef | Web of Science | Medline
To the Editor:
The study by Avidan et al. does not support routine BIS monitoring as part of standard practice, which is contrary to the study by Myles et al.1 The end-tidal anesthetic gas (ETAG)–guided protocol in the study by Avidan et al. instructed the anesthesiologists to administer a certain amount of inhaled anesthetics and use audible alarms set at the minimum level. Such an ETAG-guided protocol can be considered to be a certain intervention that can prevent awareness to some extent. An ETAG-guided protocol is not routine in clinical practice, and thus the result of the study may not favor BIS monitoring. Furthermore, given the low incidence of awareness, the sample size of this study may not have been sufficient to reach a statistical difference. In a word, the conclusion of this study should be considered very carefully.
1 ReferencesLulong Bo, M.D.
Jinbao Li, M.D.
Xiaoming Deng, M.D.
Changhai Hospital, Shanghai 200433, China1
Myles PS ,Leslie K ,McNeil J ,Forbes A ,Chan MT . Bispectral index monitoring to prevent awareness during anesthesia: the B-Aware randomised controlled trial. Lancet 2004;363:1757-1763
CrossRef | Web of Science | Medline
To the Editor:
In their double-blind, randomized, prospective, clinical trial with 2000 patients, Avidan et al. showed that the use of BIS monitoring is purposeless during the use of inhaled anesthetics for the maintenance of anesthesia. This could be speculated by most anesthesiologists, since they usually use the ETAG concentration to ensure the depth of anesthesia and prevent awareness. In another study, BIS was linearly related to the ETAG concentration.1 In the practice of anesthesia, the induction of anesthesia usually takes place with intravenous anesthetics, and in many cases, anesthesia is maintained with total intravenous anesthetics too. Currently, it is not possible to monitor the plasma concentration of the intravenous anesthetic continuously. In these cases, there may be clinical and economic benefits of BIS monitoring. In one study, the use of BIS monitoring reduced the rate of infusion of intravenous anesthetic (propofol).2 With the titrated propofol infusion, the patients were extubated significantly sooner, were more oriented on arrival in the recovery room, and were more rapidly discharged from the recovery room. Therefore, BIS monitoring is very useful for preventing awareness during total intravenous anesthesia, and it should be a part of standard practice whenever this kind of anesthesia is used, especially in high-risk patients.
2 ReferencesDiamanto Aretha, M.D.
Panagiotis Kiekkas, R.N., Ph.D.
Eleftheria Panteli, M.D.
University Hospital of Patras, 26509 Patras, Greece
adaretha@yahoo.gr 1
Katoh T ,Bito H ,Sato S . Influence of age on hypnotic requirement, bispectral index, and 95% spectral edge freguency associated with sedation by sevoflurane. Anesthesiology 2000;92:55-61
CrossRef | Web of Science | Medline2
Gan TJ ,Glass PS ,Windsor A , et al. Bispectral index monitoring allows faster emergence and improved recovery from propofol, alfentanyl, and nitrous oxide anesthesia. Anesthesiology 1997;87:808-815
CrossRef | Web of Science | Medline
To the Editor:
Avidan and colleagues report that BIS monitoring does not reduce awareness during anesthesia. More than half of the patients in the BIS group breached the protocol (i.e., BIS rose to >60). In three quarters of the patients in the control group, the prescribed limits were breached. The two groups had the same time-averaged BIS scores and average concentrations of volatile anesthetic. Since the two groups had essentially the same anesthetic, it is not surprising that the incidence of awareness was the same. A monitor will not alter the anesthesia outcome unless the information derived from it is acted on.
Notably, none of the nine patients who had definite or possible awareness received nitrous oxide as part of anesthesia. Of those who did not have awareness, 701 received nitrous oxide and 1044 did not. With the use of Fisher's exact test, the incidence of awareness with nitrous oxide is significantly lower than that without nitrous oxide (P=0.02). In a recent study of more than 200,000 cases (incidence of awareness, 1 in 14,500), none of the aware patients had received nitrous oxide.1 This finding requires further evaluation.
1 ReferencesTim M. Cook, F.R.C.A.
Royal United Hospital, Bath BA1 3NG, United Kingdom
timcook007@googlemail.com 1
Pollard RJ ,Coyle JP ,Gilbert RL ,Beck JE . Intraoperative awareness in a regional medical system: a review of 3 years' data. Anesthesiology 2007;106:269-274
CrossRef | Web of Science | Medline
Author/Editor Response
We agree with Kelley et al. that both the BIS-group and the control-group protocols may have prevented awareness in high-risk patients. It is not surprising that, despite widespread education, anesthetic dosing was not reduced by the BIS protocol, since intensive BIS training has not been shown consistently to affect anesthetic dosing1,2 or BIS readings.2
In response to Bo et al.: a major question related to our study is whether BIS monitoring or implementation of a protocol that increases vigilance is responsible for preventing awareness. For widespread adoption of BIS monitoring, the BIS protocol must be demonstrably superior to a protocol based on standard monitoring technology. Neither protocol was intended to be restrictive; practitioners were free to exercise clinical judgment. Similarly, in the study by Myles et al., 47% of patients on the BIS protocol had BIS values above 60.3 Most patients in our study had multiple risk factors, and 95% had at least one major inclusion criterion. The contention that they were at lower risk than patients in other studies is speculative, since the contribution of individual factors to overall risk has not been well studied.
The meta-analysis is invalid because the studies are not comparable for two primary reasons: the control group in the B-Aware study did not follow a protocol, and 43% of the patients received total intravenous anesthesia,3 which our study specifically excluded. In response to Aretha et al.: we believe that analogous studies comparing BIS-based protocols with protocols based on target effect-site concentrations should be conducted to evaluate the efficacy of BIS in preventing awareness during intravenous anesthesia. Cook's observation about nitrous oxide is intriguing and merits further investigation.
Concern about our study's power is misplaced. Once data have been collected, power is irrelevant and the precision of the results can best be appreciated from the 95% confidence interval.4 Moreover, our study was appropriately powered to detect a clinically meaningful5 0.9% risk reduction attributed wholly to BIS monitoring. The larger the contribution of protocol guidance to the prevention of awareness, the lower the cost-effectiveness of BIS monitoring. Our study does not exclude a small risk reduction attributable to BIS monitoring, but two follow-up studies are addressing this: a multicenter study in St. Louis, Chicago, and Winnipeg enrolling 6000 high-risk patients, and a study at the University of Michigan enrolling 30,000 patients, regardless of risk profile.
5 ReferencesMichael S. Avidan, M.B., B.Ch.
Adam C. Searleman, B.S.
Alex S. Evers, M.D.
Washington University School of Medicine, St. Louis, MO 63110 .
avidanm@wustl.edu 1
Pavlin JD ,Souter KJ ,Hong JY ,Freund PR ,Bowdle TA ,Bower JO . Effects of bispectral index monitoring on recovery from surgical anesthesia in 1,580 inpatients from an academic medical center. Anesthesiology 2005;102:566-573
CrossRef | Web of Science | Medline2
Lindholm ML ,Brudin L ,Sandin RH . Bispectral index monitoring: appreciated but does not affect drug dosing and hypnotic levels. Acta Anaesthesiol Scand 2008;52:88-94
CrossRef | Web of Science | Medline3
Myles PS ,Leslie K ,McNeil J ,Forbes A ,Chan MT . Bispectral index monitoring to prevent awareness during anaesthesia: the B-Aware randomised controlled trial. Lancet 2004;363:1757-1763
CrossRef | Web of Science | Medline4
Goodman SN ,Berlin JA . The use of predicted confidence intervals when planning experiments and the misuse of power when interpreting results. Ann Intern Med 1994;121:200-206[Erratum, Ann Intern Med 1995;122:478.]
Web of Science | Medline5
Man-Son-Hing M ,Laupacis A ,O'Rourke K , et al. Determination of the clinical importance of study results. J Gen Intern Med 2002;17:469-476
CrossRef | Web of Science | Medline
Author/Editor Response
The topic of the editorial is prevention of awareness in patients undergoing general anesthesia with the use of an inhaled anesthetic gas, not other possible applications of BIS monitoring. Thus, the suggestion by Kelley and colleagues that there is a “preponderance of clinical evidence” supporting the value of BIS monitoring is misleading. In fact, only a limited number of prospective, randomized, double-blind clinical trials (as opposed to observational studies) have investigated the effectiveness of BIS monitoring in preventing awareness.1 Furthermore, it is incorrect for Kelley et al. to claim that the study by Avidan et al. “confirmed that BIS-guided care achieves a 0.2% incidence of awareness.” The study was not designed to determine whether BIS-guided monitoring alone influenced awareness. Rather, it determined the effects of BIS monitoring when used in conjunction with an active comparator, the monitoring of ETAG. In this context, there was no additional benefit of BIS monitoring. Given that ETAG monitoring is readily available in many operating rooms in North America, any additional benefit of BIS monitoring is questionable, considering the added expense. Finally, the suggestion that inadequate vigilance or poor protocol compliance by the anesthesiologist accounts for failures of BIS monitoring assumes that simplistic protocols can be used to govern anesthesia care, which is inconsistent with clinical experience.
In response to Cook's comments: patients in the study by Avidan et al. did not undergo randomization with respect to nitrous oxide treatment. In a prospective, randomized trial of 2012 patients, the incidence of major complications was higher among those treated with nitrous oxide, and there was no reduction in the incidence of awareness.2
Given the importance of this topic, the updated meta-analysis presented by Myles and colleagues should certainly undergo an impartial peer review to consider important factors such as group heterogeneity, study quality, and the different weighting of the two major trials despite similar sample sizes.3,4 Is it appropriate to include an awareness study of 30 patients?1 The meta-analysis might also analyze and discuss the incidence of possible awareness. The patients' perceptions constitute their reality, and the incidence of possible awareness was higher than that of definite awareness in both major studies.3,4
Finally, if the calculations by Kelley and colleagues are correct, hundreds of thousands of patients would be needed to determine the usefulness of the monitor as compared with ETAG. A better approach would be to focus research on understanding the underlying disorder, and eventually bring transparent technology to bear on this serious adverse outcome.
4 ReferencesBeverly A. Orser, M.D., Ph.D.
Sunnybrook Health Science Centre, Toronto, ON M4N 3M5, Canada1
Punjasawadwong Y ,Boonjeungmonkol N ,Phongchiewboon A . Bispectral index for improving anaesthetic delivery and postoperative recovery. Cochrane Database Syst Rev 2007;4:CD003843-CD003843
Medline2
Myles PS ,Leslie K ,Chan MT , et al. Avoidance of nitrous oxide for patients undergoing major surgery: a randomized controlled trial. Anesthesiology 2007;107:221-231
CrossRef | Web of Science | Medline3
Avidan MS ,Zhang L ,Burnside BA , et al. Anesthesia awareness and the bispectral index. N Engl J Med 2008;358:1097-1108
Full Text | Web of Science | Medline4
Myles PS ,Leslie K ,McNeil J ,Forbes A ,Chan MT . Bispectral index monitoring to prevent awareness during anaesthesia: the B-Aware randomised controlled trial. Lancet 2004;363:1757-1763
CrossRef | Web of Science | Medline
- Citing Articles (1)
Citing Articles
1
David Green, Lise Paklet. (2010) Latest developments in peri-operative monitoring of the high-risk major surgery patient. International Journal of Surgery 8:2, 90-99
CrossRef
