Bispectral Index Monitoring, Duration of Bispectral Index Below 45, Patient Risk Factors, and Intermediate-term Mortality after Noncardiac Surgery in the B-Unaware Trial

• Processed electroencephalogram indices are used to assess depth of hypnosis.

• Prolonged periods of deep anesthetic levels (i.e. , bispectral index less than 45) have been associated with increased postoperative mortality.

• As a substudy of the B-Unaware Trial, the current study with noncardiac patients (N = 1,473) found no evidence that either bispectral index monitoring or avoidance of prolonged periods of low bispectral index values (i.e ., below 45) decreased intermediate-term mortality (3.2 ± 1.1 yr).

• Instead several patient characteristics were associated with increased mortality, namely: male sex, previous stroke, type II diabetes, surgery for malignancy, higher ASA (American Society of Anesthesiologists) Physical Status Classification System category, and extended intensive care unit treatment time.

THE rates of early (30 day) and intermediate term (less than 5 yr) postoperative mortality can be substantial in patients undergoing noncardiac surgery.1,2Several patient- and surgery-related factors that may impact the survival of noncardiac surgical patients have been identified during the past several decades. In contrast, less attention has been focused on anesthesia-related factors and their possible associations with short- and intermediate-term survival after noncardiac surgery. It has been suggested that, when a processed electroencephalographic index is used during general anesthesia, patients generally receive lower doses of hypnotic drugs and emerge faster from anesthesia with less postoperative nausea and vomiting.3It has also been proposed that lower doses of anesthetics could lead to a reduction in serious morbidity or mortality through avoidance of intraoperative hypotension and potential organ toxicity.4The bispectral index monitor (BIS® monitor; Covidien, Boulder, CO) is one of several candidate depth-of-anesthesia monitors based on processed electroencephalography. It has been reported that the cumulative duration of BIS values below certain arbitrary thresholds was associated with increased morbidity and intermediate-term postoperative mortality.5,6These studies have led to speculation of a mortality-hypnosis association, whereby a relative overdose of anesthetic agents causes poorer outcomes in patients with anesthetic hypersensitivity.7It has been further proposed that the outcome of surgical patients could be improved in patients receiving BIS-guided anesthesia.6,8In a planned substudy of the B-Unaware Trial in cardiac surgery patients,9we assessed relationships among cumulative duration of low BIS values (arbitrarily defined as less than 45 or 40), perioperative risk factors, anesthetic dose, and intermediate-term mortality after cardiac surgery.10Analysis using a multivariate model demonstrated that cumulative durations of BIS values at these levels were associated with a higher risk of intermediate-term mortality, independent of average end-tidal anesthetic concentration (ETAC), duration of anesthesia, or the use of BIS monitoring.10 

Parallel to our substudy in cardiac surgery patients, we undertook a similar predetermined substudy of the B-Unaware Trial in patients who underwent noncardiac surgery to further investigate relationships among perioperative factors and mortality. Through a comprehensive, multivariable model of an extensive dataset, we sought to test the hypotheses that cumulative duration of BIS values that were less than 40 or 45, cumulative anesthetic dose, comorbidities, and intraoperative events are independently associated with increased intermediate-term postoperative mortality. We specifically aimed to shed more light on the low BIS mortality hypothesis, and to question whether, in a clinically relevant range (e.g ., 0.5 to 1.5 age-adjusted minimum alveolar concentration [MAC]) increasing anesthetic exposure is dangerous.

The B-Unaware Trial was a randomized controlled trial (NCT00281489) that enrolled patients at high risk for perioperative awareness and tested whether the incidence of awareness during general anesthesia was reduced when clinicians followed a BIS-guided protocol rather than an ETAC-guided protocol.9The Human Research Protection Office at the Washington University School of Medicine (St. Louis, Missouri) approved the study, which was conducted at Barnes-Jewish Hospital (St. Louis) between September 2005 and October 2006. The study protocol had several predetermined secondary outcomes, including the aim to study the association between perioperative factors and intermediate-term mortality after surgery.

A population of 1,941 surgical patients, aged at least 18 yr, were screened and prospectively enrolled after written informed consent was obtained. Patients at high-risk for intraoperative awareness were identified based on the presence of one major criterion or two minor criteria. The major criteria were ASA (American Society of Anesthesiologists) Physical Status Classification System category P4 (systemic life-threatening disease) or P5 (not expected to survive without the operation); marginal exercise tolerance not resulting from musculoskeletal dysfunction; pulmonary hypertension; a left ventricular ejection fraction lower than 40%; aortic stenosis; pulmonary hypertension; end-stage lung disease; planned open heart surgery; preoperative long-term use of anticonvulsant agents, opiates, benzodiazepines, or cocaine; and daily alcohol consumption. The minor criteria were as follows: moderate exercise tolerance not resulting from musculoskeletal dysfunction, smoking two or more packs of cigarettes per day, chronic obstructive pulmonary disease, preoperative use of β-blockers, and obesity (body mass index [weight in kilograms divided by the square of height in meters] of more than 30 kg/m²). Of the 1,941 patients, 1,473 underwent noncardiac surgery and were included in this B-Unaware Trial substudy.

According to the original study design of the B-Unaware Trial, the only protocol-based restriction of anesthetic technique required desflurane, sevoflurane, or isoflurane as the primary agent for the maintenance of general anesthesia. When supplemental nitrous oxide was administered, it was factored into the MAC calculation. Patients were randomly assigned to a BIS-guided protocol or an ETAC-guided protocol. Practitioners in both groups could view ETACs. In the BIS group, an audible alarm was set to indicate when values exceeded 60 or fell below 40. No ETAC alarms were set in the BIS group, and practitioners were not instructed to maintain ETACs within any range. In the ETAC group, an audible alarm was set to indicate when values either fell below 0.7 MAC or exceeded 1.3 MAC. Peripheral intravenous, and, if indicated on clinical grounds, radial artery or central venous catheters were inserted. ETAC was monitored throughout the case.

A BIS® Quatro Sensor (version XP; Covidien) was applied to the forehead of each patient. BIS, hemodynamic, and anesthetic parameters were recorded at 1-s intervals and downloaded to a centralized database for subsequent analysis with TrendFace Solo software (ixellence GmbH, Wildau, Germany). For patients with incomplete datasets, manual records of anesthesia and digital photographs of monitor trends were used as alternatives.

A standard set of perioperative data were collected using information from medical records, surgical and pathology reports, anesthetic and postoperative records, discharge letters, and records of the outpatient clinic visit. All data were entered in a standard electronic data-collection form. Two investigators (M.D.K., N.P.) ascertained the quality of data collection through regular checks for completeness of the collected data and cross-checking for inconsistencies or missing information between the collected database and medical records. Aspect Medical Systems, the manufacturer of the BIS® monitor at the time the trial was conducted, had no role in the study design, data collection, data analysis, or manuscript preparation. In addition, no study monitors or other means of support were provided by Aspect Medical Systems.

According to the original study protocol of the B-Unaware Trial, patients were observed prospectively after surgery. In addition, telephone interviews were conducted at 30 days and 1 yr after surgery. The Social Security Death Index††was used to checked for vital status. This follow-up regimen was further extended with final follow-up performed in January 2010. In addition, for any patients who died after surgery, we used a variety of approaches to ascertain the cause and time of death. Thirty-day mortality and cause of death was ascertained using hospital records, discharge letters, and, whenever available, autopsy reports. For patients who survived beyond 30-day follow-up, information about the vital status was ascertained using a two-step approach. First, any subsequent hospital admissions, treatments, and follow-up visits were determined using the hospital's database. Next, the Social Security Death Index was accessed to verify vital status as of January 28, 2010.

Potential clinical determinants of intermediate-term mortality included patient characteristics, ASA Physical Status Classification System category, chronic medication use, preoperative laboratory values (hemoglobin, leukocyte count, serum creatinine). In addition, we collected information on potential intraoperative outcome determinants, including type and dose of intravenous anesthetic drugs, duration of general anesthesia, the duration of low and high mean arterial pressure, and the durations that bradycardia and tachycardia occured. Intraoperative hypotension (mean arterial pressure lower than 55 mmHg) or hypertension (mean arterial pressure higher than 100 mmHg), and intraoperative bradycardia (heart rate less than 45) and tachycardia (heart rate higher than 110) were defined according to the methodology of Reich et al  11The number of erythrocyte, fresh frozen plasma, and platelet units transfused intraoperatively were also noted. Intraoperative use of inotropes, vasopressors, and vasodilators were recorded. To study the possible association between cumulative and average ETACs of different volatile anesthetics and mortality, age-adjusted MAC values were calculated according to the recommendations published by Nickalls and Mapleson,12which include adjustments for nitrous oxide. BIS and end-tidal age-adjusted MAC values recorded at 1-s intervals were processed using MATLAB statistical software (version 7.8; The MathWorks Inc., Natick, MA) with anesthetic maintenance values defined as greater than 0.4 MAC and within ± 0.2 MAC within 3-min periods. As it was unclear whether the average anesthetic concentration or the cumulative duration of anesthetic exposure might lead to poor outcomes, average ETAC and cumulative maintenance anesthetic dose (MAC − hours = average ETAC × hours of anesthetic maintenance) were determined. Data on preexisting malignancies (for which surgery was performed) were obtained from reviewing medical records, surgical notes, and histopathologic results. Subsequently, patients were classified into subgroups according to whether the indication for surgery was performed for preexisting malignancy, and whether that malignancy was primary or a metastatic disease. Surgery type was also noted and classified into categories according to risk of mortality,2including high risk (vascular surgery; prolonged surgery [i.e ., more than 2 h], anticipated large fluid shift or blood loss [Whipple procedure, major spinal surgery]), intermediate risk (orthopedic, urologic, abdominal or thoracic, head and neck, or prostate surgery; carotid endarterectomy), and low risk (endoscopy, bronchoscopy, hysteroscopy, cystoscopy, dermatologic procedures, breast biopsy or other breast surgery, ophthalmologic procedures). In addition, the hospital's electronic database was reviewed for the length of stay in the intensive care unit.

In order to avoid misclassification of the cause of death, all-cause mortality was chosen as the study's outcome. Nevertheless, we made every effort to ascertain the cause of death through medical records and death certificates.

Descriptive statistics are presented as frequency and percentage for categorical variables and mean ± SD or median (interquartile range) for continuous variables. Comparisons were made using the Student t  test, ANOVA, Kruskal-Wallis test, or chi-square test, as appropriate.

Two recent studies reported that patients with increased cumulative duration of BIS less than 45 had an increased risk of short- and intermediate-term mortality.5,13The study of the association between cumulative duration of BIS values lower than 45 and intermediate-term mortality after noncardiac surgery was specified a priori  based on the original study design of the B-Unaware Trial. The Kaplan–Meier method was applied to evaluate the prognostic importance of the duration of BIS values less than 45 with respect to survival. Differences among survival curves were compared using the log-rank test. Univariable and multivariable Cox proportional hazards regression models were applied to evaluate relations among preoperative, intraoperative, and postoperative clinical variables, duration of BIS values less than 45, and all-cause mortality. Univariable associations to which P  values of less than 0.25 had been assigned were considered in the construction of the multivariable Cox proportional hazards regression model. The final multivariable Cox proportional hazards regression model was then derived according to the backward deletion of least-significant predictors. This statistical model is a class of survival model in which predetermined covariates or risk factors (such as patient characteristics and comorbidities) are assessed for their independent association with hazard of an event occurring or rate to a negative outcome (i.e ., mortality). Hazard ratios and corresponding 95% CIs are reported. These analyses were performed using SPSS statistical software (version 16.0; SPSS Inc., Chicago, IL). Furthermore, we quantified the discriminatory power of the final multivariable model by the c-index, which corresponds to the area under the receiver operating characteristic curve, ranging from 0.5 (performance at chance) to 1.0 (perfect performance). To evaluate the discriminatory power of the final multivariable model, a bootstrap method was used to assess the degree of over-optimism. Over-optimism occurs when application of statistical modeling techniques results in models that inaccurately predict outcomes on subsequent datasets. A bootstrapping procedure is one method which can be used to try to correct for this over-optimism.14The covariates in the final model were fitted for each bootstrap sample. The original dataset was fitted using the coefficients of the bootstrap sample model and, thus, a c-index statistic was generated from this fit on the original dataset. Optimism was then estimated as the difference in the c-index statistic from the bootstrap sample and that from the bootstrap model fit on the original sample. These differences were averaged across 100 bootstrapped samples and the difference in the original model c-index statistic and the average optimism provided the model c-index statistic corrected for optimism. These analyses were performed using the survival and design libraries within the Software R statistical environment (version 2.9.1; The R Foundation for Statistical Computing, Vienna, Austria).

There were 1,473 noncardiac surgery patients included in this study. Mean ± SD patient age was 57.9 ± 14.4 yr. Seven hundred forty-nine patients (50.8%) were men. Sixty-one percent of the enrolled study population had a history of hypertension; 27.8%, ischemic heart disease; 25.1%, chronic pulmonary disease; 19.9%, type 2 diabetes mellitus; and 14.5%, significant extracardiac arteriopathy. Eleven percent of patients had a history of previous coronary artery bypass surgery; 10.3%, previous percutaneous coronary intervention; 9.1%, chronic renal dysfunction; 8.7%, chronic congestive heart failure; and 6.2%, cerebrovascular disease. Thirty-seven percent of the cohort was classified as category P1 or P2 using the ASA Physical Status Classification System; 54.7%, P3; and 8.2%, P4. Sixty-five percent of patients underwent an operation other than surgery for cancer, 26.9% had surgery for primary cancer, and 8.6% had surgery for metastatic cancer. The majority of patients (48.1%) underwent intermediate risk surgical procedures, 27% had a high-risk surgical procedure, and 24.9% had a low-risk surgical procedure. The average duration of general anesthesia was 3.4 h (interquartile range, 2.1–4.9 h), the average intraoperative BIS value was 45.0 ± 7.8, and the average cumulative duration of BIS values lower than 45 was 1.2 h (interquartile range, 0.4–2.4 h). Baseline and clinical characteristics of the patients stratified according to the quartiles of the cumulative duration of BIS levels lower than 45 are presented in tables 1 and 2.

Mean ± SD follow-up was 3.2 ± 1.1 yr. Mortality rate was 1.5% (22 of 1,473) at 30 days and 22.8% (336 of 1,473) 3.2 yr after noncardiac surgery. The average BIS value was not significantly different between patients who were randomized to the BIS-guided protocol compared with the average BIS value among patients who were randomized to the ETAC-guided protocol (42.9 ± 8.0 vs . 43.3 ± 9.4; P = 0.50). We found no significant difference in mortality rates between patients whose general anesthesia was managed according to the BIS-guided protocol compared with patients who were managed according to the ETAC-guided protocol (24.9% [180 of 723], 95% CI, 21.8–28.1 vs . 23.7% [178 of 750], 95% CI, 20.7–26.8; P = 0.63). The causes of death are shown in table 3.

Univariable predictors of intermediate-term mortality that were significant at a nominal two-tailed P < 0.25 are shown in table 4. Many baseline characteristics as well as preoperative, intraoperative, and postoperative predictors were associated with an increased risk of intermediate-term mortality (table 4; figs. 1 and 2). There was no association between the quartiles of cumulative duration of BIS less than 45 and event-free survival reflected by the event-free survival curves (fig. 3). The results of univariable analysis showed no significant association between increased intermediate-term mortality and mean age–adjusted MAC equivalents (ETAC), and cumulative anesthetic exposure (expressed as age-adjusted MAC equivalent-hours), or with the availability of BIS-monitoring during anesthesia; and between increased intermediate-term mortality and cumulative anesthetic exposure (expressed as age-adjusted MAC equivalent-hours); or between increased intermediate-term mortality and the availability of BIS monitoring during anesthesia.

In multivariable analysis, male sex, history of stroke, type II diabetes mellitus, surgery for primary and metastatic cancer, ASA Physical Status Classification System category, and an increased duration of stay in intensive care unit were significant predictors of intermediate-term mortality (table 5). Higher body mass index and hemoglobin concentrations as well as intermediate-risk surgery were associated with decreased intermediate-term mortality. After correcting for differences in baseline characteristics as well as preoperative, intraoperative, and postoperative characteristics, there was no significant association between the cumulative duration of BIS values less than 45 and the risk of intermediate-term mortality (multivariable hazard ratio, 1.03; 95% CI, 0.93–1.14; P = 0.56).

Since the cutoff BIS number of 45 for low BIS is arbitrary and is higher than the more generally accepted arbitrary threshold of 40 for surgical anesthesia, we repeated our multivariable analysis and found that cumulative duration of BIS less than 40 (multivariable hazard ratio, 1.08; 95% CI, 0.98–1.20; P = 0.18) similarly had no significant predictive value for intermediate-term mortality. The results of the multivariable analysis also showed that allocation to the BIS-guided protocol was not associated with a decreased risk of intermediate-term mortality (multivariable hazard ratio, 1.03; 95% CI, 0.83–1.28; P = 0.78). The final multivariable model had a good c-index of 0.795, and the degree of over-optimism was minimal at 0.00413 (resulting in an adjusted c-index of 0.791). Increasing mean and cumulative ETAC were not significantly associated with intermediate-term mortality after multivariable adjustment.

This study of patients at high risk for intraoperative awareness while undergoing noncardiac surgery found no evidence that BIS monitoring or the avoidance of prolonged periods of BIS values less than 45 leads to improved intermediate-term survival. There was also no indication that higher mean or cumulative inhalational anesthetic doses, within clinically standard ranges, were associated with increased mortality. In contrast, these results suggest that specific patient characteristics and perioperative risk factors are strongly associated with intermediate-term postoperative mortality.

Several large studies have explored the association between cumulative duration of BIS values below certain arbitrary thresholds and mortality after noncardiac surgery. Monk et al .5studied 1,046 patients undergoing noncardiac surgery. Cumulative duration of BIS values lower than 45 was a significant independent predictor of 1-yr mortality. However, investigators did not stratify patients for preexisting malignant disease despite the fact that malignancy was the most common cause of death. Lindholm et al .13studied 4,087 patients undergoing noncardiac surgery who had BIS monitoring during general anesthesia. Cumulative duration of BIS values that were less than 45 was a significant predictor of 2-yr mortality, but only when preexisting malignancy status was not taken into consideration. Recently, Leslie et al .6studied 2,463 patients undergoing cardiac and noncardiac surgery who were randomized to BIS-guided anesthesia or routine care. Within the BIS group, patients were further divided into a group that had BIS values less than 40 for more than 5 min and a group that did not. The results of that study suggested that BIS monitoring and the absence of BIS values less than 40 for more than 5 min were associated with improved survival and less serious morbidity. However, BIS values were obtained only for patients randomized to BIS-guided care. In addition, investigators did not test for the effect of several important risk factors including cardiac comorbidities, surgery type, preexisting malignancy status, and the intraoperative type and dose of anesthetic drugs.

We attempted to address some of these methodological limitations in a recently published study10on a subset of 460 patients in the B-Unaware Trial who had undergone cardiac surgery. In that study investigation, we observed an association among clinical variables, intraoperative factors, and the cumulative duration of BIS values less than 45.10In contrast, there was no association between cumulative duration of BIS values lower than 45 and volatile anesthetic concentration or average total dose of intravenous drugs. These results in our companion study of cardiac surgery patients indicate that BIS values lower than 45 are likely markers of systemic illness, poor cardiac function, or complicated intraoperative course.10It is noteworthy that, in the cardiac surgery substudy, we found an independent association between cumulative duration of BIS values less than 45 and mortality.10 

In the current study, we sought to clarify further the association between clinical variables and the cumulative duration of BIS values less than 45. In addition, we sought to determine the possible role of BIS monitoring in relation to intermediate-term mortality in high-risk patients who underwent noncardiac surgery in the B-Unaware Trial. In contrast to our study of cardiac surgery patients,10the current substudy found no relationship between cumulative duration of BIS values less than 45 and mortality. The fact that an independent association between BIS values less than 45 and mortality has been found in some trials,5,10but not others,13including the current study, suggests that the association is likely epiphenomenal and, when present, is reflective of patient factors or comorbidities rather than anesthetic management. For example, in our cardiac surgery companion study, low ejection fraction was associated with cumulative duration of BIS values less than 45, whereas anesthetic dose was not.10 

The role of optimal perioperative management strategies for high-risk patients undergoing noncardiac surgery in the prevention of intermediate- and long-term mortality has been controversial. In our study, we confirmed the predictive value of many previously described risk factors for intermediate-term mortality after noncardiac surgery. Male sex,15history of stroke,2diabetes mellitus,16higher categories within the ASA Physical Status Classification System,17surgery for preexisting malignancy status,15and length of intensive care unit stay18were all significant predictors of intermediate-term mortality. The findings of our study that intermediate-risk surgery leads to lower mortality contradicts the results of previous studies and current guidelines.2,6It is likely that discrepancies arise from the differences in classification criteria; prolonged surgery for cancer is considered high-risk surgery, whereas short procedures for cancer diagnosis are classified as low risk. Low body mass index and anemia, as in previous studies, are likely to reflect the presence of general poor health or a more advanced stage of preexisting malignancy status.19–22Low body mass index has been described as a marker of patients who are more likely to have advanced disease including esophageal,20gastric,21and lung22cancers. Anemia has also been shown prospectively to be associated with increased risk for perioperative infection and adverse outcome in surgical patients.23In addition, it has been described as being associated with increased risk for hospitalization and all-cause mortality.19Many of the risk factors of poor general health including cardiovascular, metabolic, and surgery-related factors have previously been identified as modifiable predictors of long-term outcome after noncardiac surgery.16,24,25It is strategically necessary to identify and address modifiable risk factors so that both early- and long-term postoperative outcomes might be improved.

With modern multimodal anesthesia, where patients receive a combination of intravenous and varying concentrations of inhalational anesthetic agents, assessing the potential impact of anesthetic dose on mortality is challenging. We tried to address this challenge by incorporating cumulative inhalational anesthetic dose in the model, by adjusting for factors such as patient age, and by including as many anesthetic-related factors (e.g ., doses of intravenous anesthetic agents) as possible in the multivariable model. Similar to a previously published substudy10of the B-Unaware Trial, patients in the current substudy were screened and selected according to the predefined criteria of the trial, but information on some important predictors of intermediate-term mortality was not prospectively collected. Therefore, additional information was collected on perioperative risk factors using administrative data and medical records, based on physician documentation of clinical risk factors. Moreover, information was collected on preexisting malignancy, which was used to classify patients into subgroups according to whether the indication for surgery was preexisting malignancy, and whether the malignancy was primary or metastatic. Thus, the effects of some of the perioperative risk factors and preexisting malignancy status may be biased. Nevertheless, the results of the current study, similar to the findings of other studies, confirmed the predictive values of perioperative risk factors and preexisting malignancy status.13The observed mortality rates in the current study may seem higher than those reported in earlier studies.5,13However, studies with lower mortality rates may have selected patients at low risk for perioperative and late mortality. The mortality substudy of the B-Aware Trial6as well as the current substudy of the B-Unaware Trial had similarly high mortality rates, which is not surprising because both trials enrolled largely high-risk surgical patients.9,26It is noteworthy that the results of the current study do not exclude the possibility that there might be electroencephalographic features—such as prolonged or multiple episodes of burst suppression or seizure activity—during anesthesia, which are associated with increased mortality risk. The best way to investigate candidate electroencephalographic associations with adverse postoperative outcomes would be to analyze the raw electroencephalograph trace rather than a processed index, which incorporates multiple electroencephalographic components.

This study touches on the controversy of how wide the therapeutic index is for volatile anesthetics. Insufficient anesthesia is dangerous, being associated with intraoperative awareness and a high incidence of postoperative posttraumatic stress disorder.27In contrast, anesthesia dosage above a certain concentration likely becomes excessive and potentially hazardous. The question is whether, within a clinically relevant range (e.g ., 0.5–1.5 age-adjusted MAC), anesthesia is dangerous. In both B-Unaware Trial substudies,10we have tried to incorporate as many relevant potential confounding factors as possible; however, we have not found an association between higher doses of inhalational anesthetic and mortality. The previously noted association between cumulative duration of BIS values less than 45 and mortality is likely coincidental, not causal. Putting these data together, if a pharmacologically paralyzed patient were hypotensive and anesthetic was delivered at a clinically acceptable concentration (e.g ., 0.7 end-tidal age-adjusted MAC) with BIS values less than 45 (e.g ., BIS = 39), the appropriate initial intervention might not be to decrease anesthesia. Instead, it might be preferable to treat hypotension with fluids, for example, or a drug such as norepinephrine or phenylephrine.

In conclusion, this second substudy of the B-Unaware Trial found no evidence that either cumulative BIS values below a threshold of 40 or 45, or cumulative inhalational anesthetic dose is injurious to patients. In contrast, there was a strong association among perioperative risk factors, preexisting malignancy, and mortality. This study does not support the hypothesis that titrating anesthesia according to an arbitrary BIS threshold or limiting anesthetic dose would decrease intermediate-term mortality after noncardiac surgery. We do acknowledge, however, that only an appropriately designed, randomized, prospective trial might clarify further the possible presence, strength, or indeed absence of an association between a low processed electroencephalograph index and mortality.

The authors thank the many anesthesiologists, certified registered nurse anesthetists, and surgeons who facilitated this B-Unaware Trial substudy. In addition, they thank Cynthia M. Westerhout, Ph.D. (Research Associate, Division of Cardiology, Department of Medicine, University of Alberta, Edmonton, Canada), for help with statistical analysis.

Michael S. Avidan M.B.B.Ch., F.C.A.S.A. (Associate Professor of Anesthesiology, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Michael Bottros, M.D. (Resident, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Beth A. Burnside, B.A. (Research Assistant, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Alex S. Evers, M.D. (Henry E. Mallinckrodt Professor, Anesthesiology Head, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Kevin J. Finkel, M.D. (Resident, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Charles B. Hantler, M.D. (Professor of Anesthesiology, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Bernadette Henrichs Ph.D., C.R.N.A., C.C.R.N. (Director, Nurse Anesthesia Program, Goldfarb School of Nursing at Barnes-Jewish College, St. Louis, Missouri); Eric Jacobsohn, M.B.Ch.B., M.H.P.E., F.R.C.P.C. (Professor and Chair Department of Anesthesia, Department of Anesthesia and Perioperative Medicine, University of Manitoba, Winnipeg, Manitoba, Canada); Heiko Kaiser, M.D. (Research Fellow, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Miklos D. Kertai, M.D., Ph.D. (Instructor in Anesthesiology, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Andrew Leitner, B.S. (Medical Student, Washington University School of Medicine, St. Louis, Missouri); Nirvik Pal, M.D. (Instructor in Anesthesiology, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Deepika Rao, B.S (Medical Student, Washington University School of Medicine, St. Louis, Missouri); Srikar Rao, B.S. (Medical Student, Washington University School of Medicine, St. Louis, Missouri); Clare Ridley, M.D. (Resident, Department of Anesthesiology, Washington University School of Medicine, St. Louis Missouri); Leif Saager, M.D. (Instructor in Anesthesiology, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Furqaan Sadiq, B.J. (Research Assistant, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Elika Safarzadeh, M.D. (Research Assistant, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Adam C. Searleman, B.S. (Medical Student, Washington University School of Medicine, St. Louis, Missouri); Sylvia A. Searleman, B.S. (Research Assistant, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri); Jacqueline A. Selvidge, M.S. (Research Assistant, Washington University School of Medicine, St. Louis, Missouri); Brian Torres, B.S.N., R.N., C.C.R.N. (Research Assistant, Washington University School of Medicine, St. Louis, Missouri); Michelle S. Turner, M.S. (Research Assistant, Washington University School of Medicine, St. Louis, Missouri); Heidi Tymkew, M.H.S. (Research Coordinator, Washington University School of Medicine, St. Louis, Missouri); Anna Woodbury, B.S. (Medical Student, Washington University, School of Medicine, St. Louis, Missouri); Lini Zhang, M.D. (Research Assistant, Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri).