CLINICAL INVESTIGATION
BIS-guided Anesthesia Decreases Postoperative
Delirium and Cognitive Decline
Matthew T.V. Chan, MBBS, FANZCA,* Benny C.P. Cheng, MBBS, FHKCA,w
Tatia M.C. Lee, PhD,z Tony Gin, MD, FRCA, FANZCA,* and the CODA Trial Group
Background: Previous clinical trials and animal experiments
have suggested that long-lasting neurotoxicity of general anes-
thetics may lead to postoperative cognitive dysfunction
(POCD). Brain function monitoring such as the bispectral index
(BIS) facilitates anesthetic titration and has been shown to re-
duce anesthetic exposure. In a randomized controlled trial, we
tested the effect of BIS monitoring on POCD in 921 elderly
patients undergoing major noncardiac surgery.
Methods: Patients were randomly assigned to receive either BIS-
guided anesthesia or routine care. The BIS group had anesthesia
adjusted to maintain a BIS value between 40 and 60 during
maintenance of anesthesia. Routine care group had BIS meas-
ured but not revealed to attending anesthesiologists. Anesthesia
was adjusted according to traditional clinical signs and hemo-
dynamic parameters. A neuropsychology battery of tests was
administered before and at 1 week and 3 months after surgery.
Results were compared with matched control patients who did
not have surgery during the same period. Delirium was meas-
ured using the confusion assessment method criteria.
Results: The median (interquartile range) BIS values during the
maintenance period of anesthesia were significantly lower in the
control group, 36 (31 to 49), compared with the BIS-guided
group, 53 (48 to 57), P < 0.001. BIS-guided anesthesia reduced
propofol delivery by 21% and that for volatile anesthetics by
30%. There were fewer patients with delirium in the BIS group
compared with routine care (15.6% vs. 24.1%, P = 0.01). Al-
though cognitive performance was similar between groups at
1 week after surgery, patients in the BIS group had a lower rate
of POCD at 3 months compared with routine care (10.2% vs.
Received for publication June 4, 2012; accepted August 27, 2012.
From the *Department of Anaesthesia and Intensive Care, The Chinese
University of Hong Kong, Prince of Wales Hospital, Shatin;
wDepartment of Anaesthesia and Intensive Care, Tuen Mun Hos-
pital, Tuen Mun, NT; and zLaboratory of Neuropsychology, In-
stitute of Clinical Neuropsychology, The University of Hong Kong,
Pokfulam, Hong Kong Special Administrative Region.
Members of the CODA Trial Group are listed in the Appendix.
Supported by Competitive Earmarked Research Grant (CUHK4400/
06M), Research Grants Council of Hong Kong, and Health and
Health Services Research Fund (04060271).
The authors have no conflicts of interest to disclose.
Reprints: Matthew T.V. Chan, MD, MBBS, FANZCA, Department of
Anaesthesia and Intensive Care, Chinese University of Hong Kong,
Prince of Wales Hospital, Shatin, NT, Hong Kong (e-mail:
mtvchan@cuhk.edu.hk).
Copyright r 2012 by Lippincott Williams & Wilkins
J Neurosurg Anesthesiol  Volume 25, Number 1, January 2013
14.7%; adjusted odds ratio 0.67; 95% confidence interval, 0.32-
0.98; P = 0.025).
Conclusions: BIS-guided anesthesia reduced anesthetic exposure
and decreased the risk of POCD at 3 months after surgery. For
every 1000 elderly patients undergoing major surgery, anesthetic
delivery titrated to a range of BIS between 40 and 60 would
prevent 23 patients from POCD and 83 patients from delirium.
Key Words: postoperative cognitive dysfunction, depth of anes-
thesia, bispectral index, anesthetic toxicity, delirium, postoperative
complications
(J Neurosurg Anesthesiol 2013;25:33–42)
I t is widely believed that the effects of general anesthesia
are temporary and that they disappear as the drugs are
cleared from the body. There is, however, strong evidence
from animal experiments to suggest that standard doses
of routine anesthetics may produce long-lasting learning
and memory impairments that persist for weeks or
months after anesthetic exposure.1–4 This is associated
with t-hyperphosphorylation,5–7 caspase-3 activation,8–11
and b-amyloid deposition in the brain.9,12–14 Each of
these processes is directly linked to the development of
Alzheimer disease. Although human studies remain in-
conclusive,15 the current data suggest that the anesthetic
per se may be an important contributor to adverse cog-
nitive outcome after surgery.16–18
Intraoperative monitoring of processed electro-
encephalogram (EEG), such as the bispectral index (BIS),
has been shown to facilitate titration of anesthetic drug
delivery.19,20 The BIS monitor incorporates time-domain,
frequency-domain, and bispectral analysis of raw EEG
signals. This is displayed as a dimensionless number, ranges
from 0 (isoelectric EEG) to 100 (fully awake), to indicate
the depth of anesthesia.19 By aiming at a BIS value between
40 and 60 during anesthesia, the doses of hypnotic agents
were reduced by 11% to 27%.20,21 This is associated with
improved early recovery profiles and faster emergence
from anesthesia.22,23 However, it is unclear whether lower
doses of anesthetics with BIS monitoring will decrease the
risk of postoperative cognitive dysfunction (POCD). We
therefore conducted a randomized controlled trial, known as
the Cognitive Dysfunction after Anesthesia (CODA) Trial
(Centre for Clinical Trials number, CUHK_CCT00141) to
determine whether BIS-guided anesthesia decreases the
www.jnsa.com | 33
Chan et al
incidence of POCD and postoperative delirium in elderly
patients undergoing major surgery.
METHODS
Study Population
Patients were enrolled between January 2007 and
December 2009. We recruited patients who were 60 years
or older, scheduled for elective major surgery, which was
expected to last for 2 hours or longer, with an anticipated
hospital stay of at least 4 days. Patients were excluded if
they were not expected to be available for, or cooperate
with, postoperative interviews. Patients with illiteracy,
language difficulties, or significant hearing or visual im-
pairment were also excluded. Other exclusion criteria in-
cluded patients with major psychosis who were taking
tranquillizers or antidepressants, patients with diseases of
the central nervous system, suspected dementia or mem-
ory impairment with a score on the mini-mental state
examination (MMSE) of 23 or less.24
As patient performance in neuropsychology assess-
ment generally improves with repeated test admin-
istration, we recruited another 221 nonsurgical patients
from the specialist medical clinics to quantify this learning
effect. These patients underwent identical neuro-
psychological measurements during the study period.
They fulfilled the above inclusion and exclusion criteria
except they were not planned to undergo surgical proce-
dure within 6 months of enrollment.
Study Design, Randomization, and Blinding
The CODA Trial was a prospective, randomized,
double-blinded, and parallel group study. After enrollment
and immediately before induction of anesthesia, the at-
tending anesthesiologist randomized the patient according
to a computer-generated random group assigment, accessed
through an intranet system. Patients were allocated to re-
ceive either BIS-guided or routine care anesthesia. Patients,
surgeons, and all research staff, including those responsible
for postoperative data collection and outcome assessment,
were blinded to the treatment identity. The Clinical Re-
search Ethics Committee approved the study protocol, and
all patients gave written informed consents.
Study Procedure
Patients were assessed within 1 week before the
scheduled surgery. Details of medical comorbidity, surgical
history, and the number of years of education received were
recorded. Surgery, perioperative care, and safety monitor-
ing were provided according to usual local guidelines.
In the operating room, a BIS Quatro sensor (Covidien,
Mansfield, MA) was applied to the forehead of each patient,
before the induction of anesthesia according to the manu-
facturer’s recommendations. This was connected to an A-
2000 System XP monitor that was concealed from the pa-
tients and surgeons. In the BIS group, anesthetic dosage was
adjusted to achieve a BIS value between 40 and 60 from the
commencement of anesthesia to the end of surgery. An au-
dible alarm was set when the BIS number fell out of the
34 | www.jnsa.com
J Neurosurg Anesthesiol  Volume 25, Number 1, January 2013
prescribed range. In patients allocated to receive routine
care, anesthetic drug administration was titrated according
to clinical judgment. This was generally to maintain arterial
pressure within 15 mm Hg of the baseline and the heart rate
within the 40 to 90 beats/min range. If there were signs of
inadequate anesthesia, such as sweating, flushing, move-
ment, or swallowing, anesthetic dose was increased. BIS
monitoring was continued in the routine care group, but the
BIS number, its trend, and the EEG waveform were omitted
from the display, specifically designed for this trial. BIS
values, hemodynamic, and other expired gas data were re-
corded in 5-second intervals using a data acquisition pro-
gram. We calculated the time-averaged BIS value during the
maintenance of anesthesia. The amount of time when BIS
was <40 was recorded as an indicator of deep anesthesia.
In both groups of patients, cessation of general
anesthesia was timed so as to facilitate early awakening
after wound closure. We recorded the recovery times from
the end of anesthesia to eye opening and discharge from
the postanesthesia care unit (PACU) when the modified
Aldrete score was 9 or more.25 For patients who were
transferred to the intensive care unit (ICU) for post-
operative mechanical ventilation, we recorded the time to
tracheal extubation as a marker of early recovery.
After surgery, patients were regularly reviewed by
the research staff until hospital discharge. We assessed
delirium daily in the mornings after surgery using the
confusion assessment method criteria.26 Delirium was
defined as acute fluctuating course of inattention and
either disorganized thinking or an altered level of con-
sciousness. Patients who were alert were asked to rate
their quality of recovery (QoR) using the Chinese QoR
score.27 Before operation and 3 months after surgery,
patients completed a short-form health survey (SF-36) to
indicate the quality of their functional health status.
Cognitive Measurements
We measured cognitive function within a week be-
fore surgery, and again at 1 week and 3 months after
surgery. All assessments were conducted by trained re-
search staff in a quiet room.
Patients were asked to complete a Chinese version
of the cognitive failure questionnaire (CFQ) to indicate
potential subjective problems with perception, memory,
and motor function.28 A battery of 3 neuropyschological
tests was then administered to all patients.
(1) Verbal fluency test requires patients to name as many
words as possible from a predefined category (eg,
animals) within 60 seconds. We recorded the number
of words that are correctly recalled from the relevant
category.29
(2) Chinese auditory verbal learning test is a word-list
learning task that assesses verbal learning, retention,
and recognition memory.29 We recorded the total
number of recalled words (out of 15) from over 5
learning trials and those recalled 30 minutes later.
(3) Color trial making tests the psychomotor speed, and
we recorded the time taken to connect the number
and color sequences.30
r 2012 Lippincott Williams & Wilkins
J Neurosurg Anesthesiol  Volume 25, Number 1, January 2013
The tests were chosen because they were sensitive to
deficits in verbal memory, language, attention, psychomo-
tor skills, and executive functions and have been validated
for local population.29,30 To minimize learning effects, we
used parallel forms of each test and the sequence of test
administration was randomly assigned.
Change in objective cognitive function was meas-
ured by comparing the baseline performance of neuro-
psychology tests with test results obtained at 1 week and 3
months after surgery. We calculated a Z score to indicate
the standardized change in each of the neuropsychology
tests by subtracting the average learning effect measured
in the nonsurgical controls and divided by the SD in this
cohort. A large Z score indicates deterioration in a par-
ticular cognitive measure from the baseline compared
with the nonsurgical controls. We defined POCD if 2 or
more Z scores were 1.96 or greater.31,32 Similarly, sub-
jective cognitive dysfunction was defined by calculating a
Z score for the CFQ using the same approach.
Outcome
We defined the primary outcome, as POCD at 3
months after surgery. Secondary outcomes were POCD at 1
week, delirium in hospital, and the rate and QoR after an-
esthesia and surgery. We also recorded major postoperative
complications up to 3 months after surgery. Cardiac com-
plications included myocardial infarction (defined as typical
rise and fall in cardiac troponins, associated with either
ischemic symptoms, changes in electrocardiography, echo-
cardiography, coronary angiography, or pathologic find-
ings), heart failure, and thromboembolism (detected by
venography, duplex ultrasonography, ventilation-perfusion
scan, or spiral computerized tomography). Respiratory
complications included pneumonia (defined as pulmonary
infiltrates in radiologic studies, associated with fever, leuko-
cytosis, or positive culture in sputum or blood sample) and
desaturation (oxygen saturation <90% for >5 min). In-
fective complications included wound infection (defined as
purulent discharge with or without positive microbial cul-
ture) or any new infection requiring antibiotic therapy.
Statistical Analyses
Assuming the incidence of POCD in the routine care
group was 30%,33 we determined that a sample size of 450
patients per group would provide 90% power to detect an
absolute risk reduction of 15% (2-sided a = 0.05).
All patients undergoing surgery with general anes-
thesia and randomized to BIS monitoring or routine care
were considered as comprising the intention-to-treat
population for all primary and secondary analyses.
POCD as the principal outcome was compared between
groups using the Fisher exact test. Recovery times were
calculated as median time to event with interquartile
ranges (IQR) using Kaplan-Meier survival analysis and
were compared between groups by log-rank test and the
Cox proportional hazards model for possible covariate
adjustment, with assessment of the requisite proportion-
ality assumptions. Other secondary endpoints were ana-
lyzed using the Fisher exact test or w2 tests. Continuous
r 2012 Lippincott Williams & Wilkins
BIS-guided Anesthesia and POCD
variables were compared between groups using unpaired
t test. Risk factors contributing to POCD and post-
operative delirium were analyzed using logistic regression.
Only factors that scored a P value <0.20 in the univariate
analysis were incorporated in the final multivariable
model. All reported P values are 2 sided.
RESULTS
We approached 1657 elderly patients scheduled for
major surgery. After screening, 62 patients were excluded
because the preoperative MMSE was r23 points: 10 pa-
tients refused consents, 4 patients rejected the study for no
specific reason, other patients were excluded because of
their participation in other studies. A total of 921 patients
were included in the CODA Trial, of whom 462 patients
(50.2%) received BIS-guided anesthesia and 459 patients
(49.8%) were randomized to the routine care group. A total
of 85.0% and 90.7% of patients completed the 1-week and
3-month assessments, respectively (Fig. 1). Baseline char-
acteristics and surgical details of the trial participants and
were similar between study groups (Table 1).
Details on anesthetic administration are shown
in Table 2. BIS monitoring reduced end-tidal volatile
concentration by 29.7% [95% confidence intervals (CI),
25.9-32.8, P < 0.001] and estimated propofol effect site
concentration by 20.7% (95% CI, 12.1-31.9, P < 0.001).
Consequently, the average BIS value during surgery in the
BIS-guided anesthesia was higher than that in the routine
care group. The amount of time when BIS < 40 was also
lower in the BIS-monitored patients.
Primary and Secondary Outcomes
The test scores of CFQs and their performance in
neuropsychology testing at baseline, 1 week, and 3 months
after surgery are summarized in Table 3. There were fewer
patients with delirium in the BIS group compared with
routine care during the index hospital admission, with ab-
solute risk reduction 8.6% (95% CI, 3.4-13.7), but the rates
of POCD at 1 week after surgery were not different between
groups. In contrast, BIS-guided anesthesia reduced the rates
of POCD up to 3 months after surgery (Table 4). The ab-
solute risk reduction was 4.5% (95% CI, 0.25-8.9). The
number needed to treat was 23 (95% CI, 6-391). The benefit
of BIS monitoring was unaffected with a multivariable
analysis that adjusted for age, sex, education status, average
BIS value during anesthesia, and postoperative delirium
while in hospital (adjusted odds ratio 0.67; 95% CI, 0.32-
0.98; P = 0.025).
Table 5 summarizes the recovery times and post-
operative complications after BIS-guided or routine care
anesthesia. BIS monitoring shortened recovery times in
PACU. The mean differences in eye opening from cessa-
tion of anesthesia was 4.3 (95% CI, 2.7-5.8) minutes, with
hazard ratio 1.53 (95% CI, 1.32-1.79), and that for PACU
discharge, the mean difference was 12.5 (95% CI, 7.0-18)
minutes, with hazard ratio 1.40 (95% CI, 1.20-1.63).
For patients transferred to the ICU for post-
operative ventilation, BIS monitoring had no effect on
www.jnsa.com | 35
Chan et al J Neurosurg Anesthesiol  Volume 25, Number 1, January 2013

1657 Patients ≥ 60 years

for surgery ≥ 2 hours
736 excluded
660 Other research
62 MMSE ≤ 23
10 Patient refused
4 No reason stated

921Patients underwent randomization

462 assigned to BIS 459 assigned to Routine

guided anesthesia Care anesthesia

8 Surgery cancelled 4 Surgery cancelled

4 Regional only 3 Regional only
6 Died before test 4 Died before test
7 Refused testing 5 Refused testing
55 Unfit for testing 42 Unfit for testing

382 underwent 1week 401 underwent 1 week

assessments assessments

26 Died before test

2 Refused testing 22 Died before test
2 Unfit for testing 3 Unfit for testing
2 Loss to follow-up

412 underwent 423 underwent

3 months assessments 3 months assessments

FIGURE 1. Flowchart of trial enrollment. BIS indicates bispectral index; MMSE, mini-mental state examination.

time to tracheal extubation or ICU discharge, and the and 8 days (IQR, 6 to 12 d) in the routine care group,
hazard ratios (95% CI) were 1.29 (0.92-1.69) and 1.23 P = 0.98. Cardiac and respiratory complications did not
(0.94-1.62), respectively. The median length of hospital differ between groups, but the rate of postoperative in-
stay was 7 days (IQR, 5 to 10 d) in the BIS-guided group fection was significantly higher in the routine care group.
The QoR score during hospital stay and the physical
summary measure of SF-36 scale at 6 months after sur-
TABLE 1. Patient Characteristics at Entry of the Trial gery were reported better after BIS-guided anesthesia
compared with routine care.
Routine Care
BIS Group Group P
Risk Factors of POCD and Delirium
No. patients 450 452
Age (y)* 68.1 ± 8.2 67.6 ± 8.3 0.42
Independent predictors of POCD and postoperative
Male sex, no. (%) 280 (62.2) 273 (60.4) delirium are listed in Table 6 and Table 7, respectively. A
Weight (kg)* 62.0 ± 11.5 61.4 ± 10.7 0.47 total of 179 patients were found to have delirium during
ASA status, no. (%) 0.58 initial hospitalization and 104 patients had POCD at 3
1-2 373 (82.8) 382 (84.5) months after surgery. Among these factors, large doses of
3-4 76 (16.9) 70 (15.5)
Preexisting medical conditions, anesthetic, a low average BIS value during surgery, long
no. (%)z period of deep anesthesia (BIS < 40), and increasing age
Cardiovascular 374 (83.1) 330 (73.0) 0.54 remained significant in a multivariable analysis. We were
Respiratory 75 (16.7) 67 (14.8) 0.70 unable to demonstrate a relationship between post-
Endocrine 109 (24.2) 101 (22.3) 0.53
Others 80 (17.8) 87 (19.2) 0.63
operative infection and POCD, but our analysis is prob-
Surgery for cancer, no. (%) 338 (75.1) 352 (77.9) 0.39 ably underpowered to detect the modest correlation.
Duration of anesthesia (h)* 2.1 ± 1.0 2.0 ± 1.1 0.67 Patients with delirium during hospitalization were
Years of education receivedw 6 (0-22) 6 (0-18) 0.84 more likely to perform poorly in the neuropsychology tests
Chinese Geriatric Depression 2 (0-13) 2 (0-14) 0.18 during the first week and 3 months after surgery, odds ratio
Scalew
Mini-mental state examination 28 (24-30) 28 (24-30) 0.64 (95% CI): 16.7 (8.7-32.1) and 12.3 (7.5-20.0), respectively.
scorew
Values are number (%). DISCUSSION
*Mean ± SD.
wMedian (range). Principal Findings
zA patient could have >1 preexisting medical conditions.
BIS indicates bispectral index. Our study demonstrated that in elderly patients
undergoing major surgery, there was 21% decrease in the

36 | www.jnsa.com r 2012 Lippincott Williams & Wilkins

J Neurosurg Anesthesiol  Volume 25, Number 1, January 2013 BIS-guided Anesthesia and POCD

TABLE 2. Comparison of Anesthetic Techniques

BIS Group Routine Care Group P
No. patients 450 452
Propofol dose (mg) 136 ± 30 148 ± 33 0.64
Opioid dose
Fentanyl (mg/kg)z 1.6 ± 0.4 1.6 ± 0.3 0.75
No. (%) 400 (88.8) 411 (90.9)
Morphine (mg)z 0.12 ± 0.07 0.12 ± 0.06 0.20
No. (%) 373 (83.0) 381 (84.4)
Midazolam dose (mg)z 2.5 ± 1.1 2.9 ± 1.4 0.76
No. (%) 33 (7.3) 27 (5.9%)
Estimated effect site propofol concentration (mg/mL) 2.7 ± 0.9 3.3 ± 1.1 <0.001
No. (%) 45 (10.0%) 54 (11.9%) 0.06
End-tidal volatile concentration (MAC equivalents)zy 0.57 ± 0.29 0.93 ± 0.34 <0.001
No. (%) 405 (90.0%) 398 (88.1%) 0.06
Nitrous oxide use
No. (%) 241 (53.5) 259 (57.4)
End-tidal concentration (%)z 63 ± 10 65 ± 10 0.70
Time average BIS valuesz 53.2 ± 8.9 38.6 ± 6.5 <0.001
Median (interquartile range) 53 (48-57) 36 (31-49)
Time when BIS < 40 (min)z 7.2 ± 7.8 22.8 ± 7.3 <0.001
Median (interquartile range) 6.9 (4.4-9.8) 25.7 (7.7-55.8)
Clinically significant hypotension*
No. (%) 53 (11.7) 61 (14.0) 0.50
Clinically significant bradycardiaw
No. (%) 17 (3.7%) 20 (4.4%) 0.11
Values are number (%).
*Clinically significant hypotension was defined as systolic arterial pressure <90 mm Hg for >15 minutes that required fluid resuscitation, an
inotropic agent, or vasopressor.
wClinically significant bradycardia was defined as heart rate <40 beats/min for >15 minutes that required sympathomimetic agent or atropine.
zMean ± SDs.
yAdjusted for age and concentration of nitrous oxide administered.
BIS indicates bispectral index; MAC, minimum alveolar concentration.

propofol delivery and 30% decrease in the administration
of volatile anesthetic when BIS was maintained between
40 and 60 during surgery. BIS-guided anesthesia also re-
duced the risk of postoperative delirium during initial
hospitalization by 35% and POCD at 3 months after
surgery by 31%. Patients receiving BIS monitoring re-
covered from anesthesia more quickly than routine care,
with earlier eye opening and faster discharge from PACU.
BIS-guided anesthesia also significantly decreased the risk
of postoperative infection, but our multivariable analysis
could not demonstrate a measurable relationship between
the occurrence of infection and POCD.
Our Study in Relation to Previous Data
Two randomized trials have evaluated the effect of
anesthetic depth on POCD. In contrast to our findings,

TABLE 3. Performance in Neuropsychology Tests

Baseline 1 wk After Surgery 3 mo After Surgery
Nonsurgical BIS Routine Nonsurgical BIS Routine Nonsurgical BIS Routine
Test Controls Group Care Group Controls Group Care Group Controls Group Care Group
No. patients 221 450 452 221 382 401 221 412 423
Cognitive failure 18.9 ± 12.7 19.1 ± 13.4 19.0 ± 13.2 19.0 ± 10.7 18.7 ± 9.8 21.2 ± 12.0 19.1 ± 14.1 21.2 ± 14.5 21.3 ± 14.5
questionnaire
Verbal Fluency Test
No. correct items 12.2 ± 14.1 13.0 ± 4.1 12.4 ± 3.6 13.1 ± 4.0 12.1 ± 3.4 12.4 ± 4.0 13.1 ± 3.9 13.9 ± 15.4 12.4 ± 3.9
Chinese Auditory
Verbal Learning Test
Recall 36.7 ± 9.6 32.7 ± 8.1 33.1 ± 8.9 45.4 ± 15.7 42.8 ± 16.2 43.1 ± 14.6 42.2 ± 17.3 40.0 ± 15.4 37.4 ± 14.9
(no. words)
Delay recall 6.4 ± 3.5 5.9 ± 34 6.1 ± 2.9 9.3 ± 3.4 8.4 ± 3.7 8.5 ± 3.3 9.3 ± 3.2 8.3 ± 3.5 7.8 ± 3.4
(no. words)
Color Trial Making
Reaction time (s) 71.1 ± 32.6 63.8 ± 29.9 64.6 ± 32.2 54.6 ± 90.9 64.2 ± 36.8 66.9 ± 33.2 62.1 ± 30.2 60.5 ± 28.8 62.3 ± 44.4
Values are mean ± SD.
BIS indicates bispectral index.

r 2012 Lippincott Williams & Wilkins www.jnsa.com | 37

Chan et al J Neurosurg Anesthesiol  Volume 25, Number 1, January 2013

TABLE 4. Postoperative Cognitive Outcomes

BIS Group Routine Care Group Odds Ratio (95% CI) P
Cognitive failure questionnaire at 3 mo after surgery
No./total no. (%) 29/412 (7.0%) 31/423 (7.3%) 0.95 (0.41-1.98) 0.14
Delirium
No/total no. (%) 70/450 (15.6%) 109/452 (24.1%) 0.58 (0.41-0.80) 0.01
Postoperative cognitive dysfunction
1 wk after surgery
No./total no. (%) 83/382 (21.7%) 93/401 (23.1%) 0.92 (0.66-1.29) 0.06
3 mo after surgery
No./total no. (%) 42/412 (10.2%) 62/423 (14.7%) 0.62 (0.39-0.97) 0.02
BIS indicates bispectral index.

Farag et al34 reported better cognitive function with
higher processing index in 74 patients receiving larger
doses of anesthetics (BIS, 30 to 40) after noncardiac
surgery. Similarly, An et al35 reported a lower incidence
of POCD in 80 patients undergoing craniotomy for mi-
crovascular decompression when BIS was maintained in
the range of 30 to 40 (deeper anesthesia) compared with
those in the lighter anesthesia group (BIS, 55 to 65).
However, both studies focused on cognitive measure-
ments in the early postoperative period. Given that cog-
nitive assessments during this period are greatly
influenced by postoperative pain, analgesic therapy, and
physical disability, it is unclear that how these results
might reflect the impact of anesthesia on POCD. Inter-
estingly, in an observational study of 65 elderly patients
having noncardiac surgery, anesthetic depth, measured by
the median (5% to 95% percentiles) cerebral state EEG
index, was similar in patients with POCD at 1 week after
surgery [40 (32 to 55); n = 9] compared with those who
did not [43 (30 to 70); n = 56], P = 0.41.36 In this regard,
the consensus statement on the assessment of neuro-
behavioral outcomes after cardiac surgery recommended
that cognitive assessment should be performed at least 6
weeks after surgery to indicate nonsurgical insults to the
brain.32 In the CODA Trial, we did not observe difference
in cognitive performance between groups until 3 months
after surgery. Nevertheless, it has been postulated that
deep anesthesia, by reducing cerebral metabolism and
stress response to surgery, may decrease POCD. In ani-
mal experiments, mice receiving a higher concentration of
isoflurane demonstrated better performance in behavioral
tests,37,38 but deep anesthesia with large doses of propofol
has been associated with higher incidence of neurological
deficits after cardiac surgery.39
Two large cohort studies have suggested potential
harmful effects of deep anesthesia. Monk et al40 studied
the outcome in 1046 patients after noncardiac surgery.
Deep anesthesia, defined as cumulative time when
BIS < 45, significantly predicted 12-month mortality,
with relative risk 1.24 (95% CI, 1.06-1.44; P = 0.012).

TABLE 5. Recovery Profiles and Postoperative Complications

BIS Group Routine Care Group P
No. patients 450 452
PACU admission, no.(%) 336 (74.7) 332 (73.4) 0.56
Time to eye opening (min)* 10 (7-15) 15 (10-21) <0.001w
Time to discharge from PACU (min)* 80 (65-105) 92 (70-120) <0.001w
Intensive care unit admission, no.(%) 113 (25.1) 120 (26.6) 0.56
Time to tracheal extubation (h)* 0.4 (0.2-4.5) 2.5 (0.2-7.9) 0.91w
Intensive care unit stay (d)* 2.0 (1.9-2.1) 2.0 (1.8-2.1) 0.06w
Hospital stay (d)* 7 (5-10) 8 (6-12) 0.98w
Quality of recovery score (out of 18)
Day 1 11.8 ± 2.1 9.8 ± 2.4 <0.001
Hospital discharge 16.3 ± 1.7 15.3 ± 2.1 <0.001
Short-form health survey SF-36 at 3 mo after surgery:
Physical summary measures 47.4 ± 9.2 45.1 ± 10.2 0.002
Mental summary measures 50.2 ± 12.1 52.1 ± 10.9 0.053
Postoperative complications, no.(%)
Cardiac 28 (6.2) 33 (7.3) 0.13
Respiratory 64 (14.2) 81 (17.9) 0.67
Infection 75 (16.7) 104 (23.0) 0.02
Any complication 48 (10.7) 94 (20.8) 0.01
Values are number (%) or mean ± SD.
*Median (interquartile range).
wLog-rank test.
BIS indicates bispectral index; PACU, postanesthesia care unit.

38 | www.jnsa.com r 2012 Lippincott Williams & Wilkins

J Neurosurg Anesthesiol  Volume 25, Number 1, January 2013 BIS-guided Anesthesia and POCD

TABLE 6. Risk Factors of Cognitive Dysfunction at 3 Months After Surgery (n = 835)

Univariate Analysis Multivariable Analysis
Risk Factors Summary Statistics* Odds Ratio P Odds Ratio P
Age (y)
No POCD 67 ± 7.9
POCD 68 ± 8.9 1.02 (1.00-1.04) 0.02 1.04 (1.01-1.08) 0.01
Education (y)
No POCD 7.2 ± 4.0
POCD 6.1 ± 4.2 0.94 (0.88-0.99) 0.04
Male sex (n = 475), no. (%)
No POCD 453 (62.1)
POCD 62 (58.8) 1.15 (0.73-1.80) 0.25
Postoperative infection, no. (%)
No POCD 114 (15.6)
POCD 33 (31.7) 2.17 (1.50-3.15) 0.001
Respiratory complication, no. (%)
No POCD 86 (11.8)
POCD 20 (18.5) 1.69 (1.01-2.89) 0.02
Delirium, no. (%)
No POCD 94 (12.9)
POCD 68 (64.4) 12.26 (7.50-20.0) <0.001 9.58 (4.62-19.9) <0.001
Intraoperative BIS value
No POCD 48.2 ± 6.2
POCD 37.4 ± 3.9 0.65 (0.59-0.70) <0.001 0.93 (0.85-0.97) <0.001
Time with BIS < 40 (h)
No POCD 0.3 ± 0.5
POCD 0.7 ± 0.6 1.21 (1.01-1.30) 0.03 1.11 (1.01-1.96) 0.04
End-tidal volatile concentration (MAC equivalents)
No POCD 0.74 ± 0.25
POCD 1.16 ± 0.31 8.48 (5.15-13.97) <0.001 2.31 (1.15-15.6) 0.03
*Values are number (%) or mean ± SD.
BIS indicates bispectral index; MAC, minimum alveolar concentration; POCD, postoperative cognitive dysfunction.

TABLE 7. Risk Factors of Postoperative Delirium (n = 902)

Univariate Analysis Multivariable Analysis
Risk Factors Summary Statistics* Odds Ratio P Odds Ratio P
Age (y)
No delirium 68 ± 8.1
Delirium 69 ± 9.0 1.03 (1.00-1.05) 0.05
Education (y)
No delirium 6.9 ± 4.1
Delirium 6.4 ± 4.3 0.97 (0.94-1.01) 0.06
Male sex (n = 475), no. (%)
No delirium 443 (61.1)
Delirium 113 (64.2) 0.88 (0.63-1.24) 0.48
Postoperative infection, no. (%)
No delirium 128 (17.7)
Delirium 51 (28.4) 1.86 (1.28-2.71) 0.001
Respiratory complication, no. (%)
No delirium 108 (14.9)
Delirium 37 (20.8) 1.49 (0.98-2.26) 0.06
Intraoperative BIS value
No delirium 47.8 ± 6.5
Delirium 37.7 ± 7.3 0.87 (0.84-0.89) <0.001 0.91 (0.87-0.96) <0.001
Time with BIS < 40 h
No delirium 0.3 ± 0.4
Delirium 0.8 ± 0.5 3.70 (2.00-8.11) <0.001 2.05 (1.02-4.16) 0.03
End-tidal volatile concentration (MAC equivalents)
No delirium 0.76 ± 0.27
Delirium 1.01 ± 0.34 6.81 (2.21-22.7) <0.001 1.15 (1.05-7.34) 0.04
*Values are number (%) or mean ± SD.
BIS indicates bispectral index; MAC, minimum alveolar concentration.

r 2012 Lippincott Williams & Wilkins www.jnsa.com | 39

Chan et al
Similarly, Lindholm et al41 found an association between
deep anesthesia, using the same definition, and 2-year
mortality in 4087 patients having noncardiac surgery with
intraoperative BIS monitoring, but this was only sig-
nificant when preexisting malignancy was included, with
hazard ratio 1.18 (95% CI, 1.08-1.29; P = 0.003). Al-
though these findings were intriguing, the 2 studies have
been criticized for the observational cohort design and
potential bias.42
The B-Aware Trial has recently reported a higher
mortality with deep anesthesia, defined as BIS < 40 for
>5 minutes, during a median follow-up of 4.1 (range, 0
to 6.5) years after surgery, with hazard ratio 1.42 (95%
CI, 1.04-1.93; P = 0.03).43 Episodes of deep anesthesia
were also associated with an increased risk of myocardial
infarction and stroke during the follow-up, with hazard
ratio 1.94 (95% CI, 1.12-3.35; P = 0.02) and 3.23 (95%
CI, 1.29-8.07; P = 0.01), respectively. Although the
B-Aware Trial was a large randomized study, the ob-
servations could not be considered as conclusive, because
BIS data were obtained in only half of the patients
(BIS-guided group) and that only intermittent BIS values
were recorded. Furthermore, some of the important pre-
dictors for postoperative morbidity and mortality, such as
cancer status, have not been included in the analysis.43,44
Interestingly, in a post hoc analysis of another large
randomized study, the B-Unaware Trial, cumulative time
with BIS < 45 was predictive of mortality after cardiac
surgery, with hazard ratio 1.29 (95% CI, 1.12-1.49),45 but
not with noncardiac procedures, with hazard ratio 1.03
(95% CI, 0.93-1.14).46
More recently, 2 retrospective analyses of large an-
esthetic databases have suggested that a low BIS value in
combination of low arterial pressure during low anesthetic
delivery (triple low) was associated with 2.5- to 4-fold
increase in mortality.47,48 This scenario suggested an in-
creased sensitivity to anesthetic and may be a marker of
underlying organ dysfunction and hypoperfusion. In this
setting, a modest dose of anesthetic could result in relative
drug overdose and may lead to poor outcome. The large
body of evidence therefore suggested that deep anesthesia
(absolute or relative) may contribute to adverse post-
operative outcomes; however, a randomized controlled trial
comparing 2 distinct levels of anesthetic depth will be re-
quired to establish the causal relationship. In the CODA
Trial, we did not randomize patients to 2 levels of anesthetic
depths, but we were successful to achieve a separation in BIS
values and anesthetic dosage between groups. Given that
patients in the routine care group, who received larger doses
of anesthetics with lower BIS values, had poorer cognitive
outcome, our data demonstrated that careful titration of
anesthetics to avoid deep anesthesia produced long-term
benefit in preventing POCD. The effect of deep anesthesia on
other rarer postoperative complications, such as death, will
require further studies. In collaboration with the Australian
and New Zealand College of Anaesthetists Trials Group, we
have started recruiting patients for the Balanced Anesthesia
Trial (Australian New Zealand Clinical Trials Registry No:
ACTRN12612000632897) to determine the impact of light
40 | www.jnsa.com
J Neurosurg Anesthesiol  Volume 25, Number 1, January 2013
versus deep general anesthesia on all-cause mortality at
one year postoperatively, in 6500 moderate to high risk
patients having major noncardiac surgery.
We found a substantial reduction in anesthetic ex-
posure with BIS monitoring compared with routine care
anesthesia. Our data are consistent with previous studies
and meta-analyses.20,21,23 However, recent large effec-
tiveness trials, the B-Unaware Trial and the subsequent
BAG-RECALL Trial,49,50 showed no change in the an-
esthetic administration with BIS-guided anesthesia. It
should be noted that both trials included an active com-
parator group, where anesthesia was titrated to maintain
an age-adjusted minimum alveolar concentration
(MAC)Z0.7. It is plausible that anesthesiologists may
provide extraordinary vigilance to prevent unintentional
awareness with the lowest possible anesthetic delivery in
the comparator group. The specific study design could
diminish the difference in anesthetic exposure between
groups. Interestingly, in the B-Aware Trial, using routine
anesthetic care as the control group, BIS monitoring re-
duced the estimated plasma concentration of propofol by
17% and dose of midazolam used by 20%.44
There are limitations with BIS monitoring during
clinical anesthesia. A number of environmental and phys-
iological factors may affect BIS performance. Electrical
50 Hz mains interference, electrocardiographic, electro-
myographic, and electrocautery artifacts introduce high-
frequency signals and are the major source of errors.51,52
Unless these factors are carefully corrected, serious mis-
interpretation can occur.51 Nevertheless, despite the inter-
individual variability, BIS monitoring seems to provide
useful information to track the anesthetic drug effect.51
Postoperative Delirium and POCD
We found that postoperative delirium was common
(20%) after major surgery in the elderly and this is con-
sistent with previous studies.53,54 More importantly, we
noted that the risk factors for postoperative delirium were
similar to those predicting POCD. This finding would
suggest that the 2 adverse outcomes may have derived
from a common mechanism, such as deep anesthesia. In
this regard, our study showed that by limiting anesthetic
exposure, there was a significant decrease in postoperative
delirium and POCD. This should facilitate short-term
rehabilitation and long-term functional recovery.
Strengths and Weakness of Our Study
Although we designed the study according to the
published guidelines,31,32 the test scores of our neuro-
psychology battery cannot be directly compared with
other studies, because different tests were used. Un-
fortunately, universal neuropsychology tests are currently
unavailable. Nonetheless, we chose tests that are sensitive
and culturally adoptable for the local population. We also
included nonsurgical controls to adjust for the practice
effect of repeated testing. Given that the incidence of
postoperative delirium and POCD are comparable with
other studies,31–33 we believe our methods are robust in
detecting adverse cognitive events.
r 2012 Lippincott Williams & Wilkins
J Neurosurg Anesthesiol  Volume 25, Number 1, January 2013
Our study population was restricted to the elderly
patients undergoing major surgery; therefore, the results
may not be generalized to patients undergoing minor
surgery with duration <2 hours. This is particularly im-
portant for the younger patients who appear to have less
risk of developing POCD,55 and so, deep anesthesia may
be less critical compared with the elderly.
Interpretation
The CODA Trial indicated that for every 1000 pa-
tients undergoing major surgery, BIS-guided anesthesia
prevented 83 patients from suffering delirium during hos-
pital admission and 23 patients from POCD at 3 months
after surgery. Given that intraoperative low BIS value, long
period of deep anesthesia (BIS < 40), and large doses of
anesthetic were predictors of POCD, BIS monitoring with
careful titration of anesthetics should prevent unintentional
deep anesthesia and may be useful for improving post-
operative cognitive performance in the elderly.
CONCLUSIONS
Our study demonstrated that in elderly patients un-
dergoing major surgery, titrating anesthetic to maintain a
BIS value between 40 and 60 during surgery avoided epi-
sodes of deep anesthesia. This was associated with a reduc-
tion in the risk of developing delirium during initial
hospitalization and POCD at 3 months after surgery. BIS
monitoring also expedite recovery from anesthesia.
REFERENCES
1. Culley DJ, Baxter M, Yukhananov R, et al. The memory effects of
general anesthesia persist for weeks in young and aged rats. Anesth
Analg. 2003;96:1004–1009.
2. Culley DJ, Baxter MG, Yukhananov R, et al. Long-term impair-
ment of acquisition of a spatial memory task following isoflurane-
nitrous oxide anesthesia in rats. Anesthesiology. 2004;100:309–314.
3. Bianchi SL, Tran T, Liu C, et al. Brain and behavior changes in
12-month-old Tg2576 and nontransgenic mice exposed to anes-
thetics. Neurobiol Aging. 2008;29:1002–1010.
4. Wan Y, Xu J, Ma D, et al. Postoperative impairment of cognitive
function in rats: a possible role for cytokine-mediated inflammation
in the hippocampus. Anesthesiology. 2007;106:436–443.
5. Ikeda Y, Ishiguro K, Fujita SC. Ether stress-induced Alzheimer-like
tau phosphorylation in the normal mouse brain. FEBS Lett.
2007;581:891–897.
6. Planel E, Richter KE, Nolan CE, et al. Anesthesia leads to tau
hyperphosphorylation through inhibition of phosphatase activity by
hypothermia. J Neurosci. 2007;27:3090–3097.
7. Run X, Liang Z, Zhang L, et al. Anesthesia induces phosphor-
ylation of tau. J Alzheimers Dis. 2009;16:619–626.
8. Fütterer CD, Maurer MH, Schmitt A, et al. Alterations in rat brain
proteins after desflurane anesthesia. Anesthesiology. 2004;100:302–308.
9. Zhang B, Dong Y, Zhang G, et al. The inhalation anesthetic
desflurane induces caspase activation and increases amyloid beta-
protein levels under hypoxic conditions. J Biol Chem. 2008;283:
11866–11875.
10. Zhang G, Dong Y, Zhang B, et al. Isoflurane-induced caspase-3
activation is dependent on cytosolic calcium and can be attenuated
by memantine. J Neurosci. 2008;28:4551–4560.
11. Kalenka A, Gross B, Maurer MH, et al. Isoflurane anesthesia elicits
protein pattern changes in rat hippocampus. J Neurosurg Anes-
thesiol. 2010;22:144–154.
12. Xie Z, Dong Y, Maeda U, et al. The common inhalation anesthetic
isoflurane induces apoptosis and increases amyloid beta protein
levels. Anesthesiology. 2006;104:988–994.
r 2012 Lippincott Williams & Wilkins
BIS-guided Anesthesia and POCD
13. Abramov E, Dolev I, Fogel H, et al. Amyloid-beta as a positive
endogenous regulator of release probability at hippocampal
synapses. Nat Neurosci. 2009;12:1567–1576.
14. Dong Y, Zhang G, Zhang B, et al. The common inhalational
anesthetic sevoflurane induces apoptosis and increases beta-amyloid
protein levels. Arch Neurol. 2009;66:620–631.
15. Cottrell JE, Hartung J. Developmental disability in the young and
postoperative cognitive dysfunction in the elderly after anesthesia
and surgery: do data justify changing clinical practice? Mt Sinai J
Med. 2012;79:75–94.
16. Xie Z, Tanzi RE. Alzheimer’s disease and post-operative cognitive
dysfunction. Exp Gerontol. 2006;41:346–359.
17. Baranov D, Bickler PE, Crosby GJ, et al. Consensus statement: First
International Workshop on Anesthetics and Alzheimer’s disease.
Anesth Analg. 2009;108:1627–1630.
18. Tang J, Eckenhoff MF, Eckenhoff RG. Anesthesia and the
old brain. Anesth Analg. 2010;110:421–426.
19. Chan MT, Gin T. What does the bispectral EEG index monitor? Eur
J Anaesthesiol. 2000;17:146–148.
20. Liu SS. Effects of Bispectral Index monitoring on ambulatory
anesthesia: a meta-analysis of randomized controlled trials and a
cost analysis. Anesthesiology. 2004;101:311–315.
21. Punjasawadwong Y, Boonjeungmonkol N, Phongchiewboon A.
Bispectral index for improving anaesthetic delivery and postoper-
ative recovery. Cochrane Database Syst Rev. 2007;CD003843.
22. Leslie K, Myles PS, Chan MT, et al. Risk factors for severe
postoperative nausea and vomiting in a randomized trial of nitrous
oxide-based vs nitrous oxide-free anaesthesia. Br J Anaesth.
2008;101:498–505.
23. Leslie K, Myles PS, Forbes A, et al. Recovery from bispectral index-
guided anaesthesia in a large randomized controlled trial of pa-
tients at high risk of awareness. Anaesth Intensive Care. 2005;33:
443–451.
24. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A
practical method for grading the cognitive state of patients for the
clinician. J Psychiatr Res. 1975;12:189–198.
25. Aldrete JA, Kroulik D. A postanesthetic recovery score. Anesth
Analg. 1970;49:924–934.
26. Inouye SK, van Dyck CH, Alessi CA, et al. Clarifying confusion: the
confusion assessment method. A new method for detection of
delirium. Ann Int Med. 1990;113:941–948.
27. Chan MT, Lo CC, Lok CK, et al. Psychometric testing of the Chinese
quality of recovery score. Anesth Analg. 2008;107:1189–1195.
28. Chan RCK. What does cognitive failure questionnaire measure?
General cognitive failure or specific domain deficits? Arch Clin
Neuropsychol. 1999;14:735–736.
29. Lee TMC, Yuen KSL, Chan CCH. Normative data for neu-
ropsychological measures of fluency, attention, and memory
measures for Hong Kong Chinese. J Clin Exp Neuropsychol. 2002;
24:615–632.
30. Lee TM, Cheung CC, Chan JK, et al. Trail making across
languages. J Clin Exp Neuropsychol. 2000;22:772–778.
31. Rasmussen LS, Larsen K, Houx P, et al. The assessment of
postoperative cognitive function. Acta Anaesthesiol Scand.
2001;45:275–289.
32. Murkin JM, Newman SP, Stump DA, et al. Statement of consensus
on assessment of neurobehavioral outcomes after cardiac surgery.
Ann Thorac Surg. 1995;59:1289–1295.
33. Monk TG, Weldon BC, Garvan CW, et al. Predictors of cognitive
dysfunction after major noncardiac surgery. Anesthesiology.
2008;108:18–30.
34. Farag E, Chelune GJ, Schubert A, et al. Is depth of anesthesia, as
assessed by the bispectral index, related to postoperative cognitive
dysfunction and recovery? Anesth Analg. 2006;103:633–640.
35. An J, Fang Q, Huang C, et al. Deeper total intravenous anesthesia
reduced the incidence of early postoperative cognitive dysfunction
after microvascular decompression for facial spasm. J Neurosurg
Anesthesiol. 2011;23:12–17.
36. Steinmetz J, Funder KS, Dahl BT, et al. Depth of anaesthesia and
post-operative cognitive dysfunction. Acta Anaesthesiol Scand. 2010;
54:162–168.
www.jnsa.com | 41
Chan et al
37. Valentim AM, Di Giminiani P, Ribeiro PO, et al. Lower isoflurane
concentration affects spatial learning and neurodegeneration in
adult mice compared with higher concentrations. Anesthesiology.
2010;113:1099–1108.
38. Valentim AM, Alves HC, Silva AM, et al. Effects of depth of
isoflurane anaesthesia on a cognition task in mice. Br J Anaesth.
2008;101:434–435.
39. Roach GW, Newman MF, Murkin JM, et al. Ineffectiveness of
burst suppression therapy in mitigating perioperative cerebrovas-
cular dysfunction. Multicenter Study of Perioperative Ischemia
(McSPI) Research Group. Anesthesiology. 1999;90:1255–1264.
40. Monk TG, Saini V, Weldon BC, et al. Anesthetic management and
one-year mortality after noncardiac surgery. Anesth Analg. 2005;
100:4–10.
41. Lindholm ML, Traff S, Granath F, et al. Mortality within 2 years
after surgery in relation to low intraoperative bispectral index values
and preexisting malignant disease. Anesth Analg. 2009;108:508–512.
42. Cohen NH. Anesthetic depth is not (yet) a predictor of mortality!
Anesth Analg. 2005;100:1–3.
43. Leslie K, Myles PS, Forbes A, et al. The effect of bispectral index
monitoring on long-term survival in the B-aware trial. Anesth Analg.
2010;110:816–822.
44. Myles PS, Leslie K, McNeil J, et al. Bispectral index monitoring to
prevent awareness during anaesthesia: the B-Aware randomised
controlled trial. Lancet. 2004;363:1757–1763.
45. Kertai MD, Pal N, Palanca BJ, et al. Association of perioperative
risk factors and cumulative duration of low bispectral index with
intermediate-term mortality after cardiac surgery in the B-
Unaware Trial. Anesthesiology. 2010;112:1116–1127.
46. Kertai MD, Palanca BJ, Pal N, et al. Bispectral index monitoring,
duration of bispectral index below 45, patient risk factors, and
intermediate-term mortality after noncardiac surgery in the B-
Unaware Trial. Anesthesiology. 2011;114:545–556.
47. Sessler DI, Sigl JC, Kelley SD, et al. Hospital stay and mortality are
increased in patients having a “triple low” of low blood pressure,
42 | www.jnsa.com
J Neurosurg Anesthesiol  Volume 25, Number 1, January 2013
low bispectral index, and low minimum alveolar concentration of
volatile anesthesia. Anesthesiology. 2012;116:1195–1203.
48. Gan TJ, White W, Hale B, et al. Mortality at one year is increased
by a “triple low” combination of BIS, blood pressure and anesthetic
concentration. Anesthesiology. 2011: A1574.
49. Avidan MS, Jacobsohn E, Glick D, et al. Prevention of intra-
operative awareness in a high-risk surgical population. N Engl J
Med. 2011;365:591–600.
50. Avidan MS, Zhang L, Burnside BA, et al. Anesthesia awareness and
the bispectral index. N Engl J Med. 2008;358:1097–1108.
51. Lobo FA, Schraag S. Limitations of anaesthesia depth monitoring.
Curr Opin Anaesthesiol. 2011;24:657–664.
52. Chan MT, Ho SS, Gin T. Performance of the bispectral index during
electrocautery. J Neurosurg Anesthesiol. 2012;24:9–13.
53. Rudolph JL, Marcantonio ER, Culley DJ, et al. Delirium is
associated with early postoperative cognitive dysfunction. Anaes-
thesia. 2008;63:941–947.
54. Saczynski JS, Marcantonio ER, Quach L, et al. Cognitive trajectories
after postoperative delirium. N Engl J Med. 2012;367:30–39.
55. Johnson T, Monk T, Rasmussen LS, et al. Postoperative cognitive
dysfunction in middle-aged patients. Anesthesiology. 2002;96:1351–1357.
APPENDIX
The CODA Trial Group:
Prince of Wales Hospital, The Chinese University of
Hong Kong: Matthew T.V. Chan, Tony Gin,
Emily G.Y. Koo, Qinzhou Wang, Keung-Tat Lee.
Tuen Mun Hospital: Benny C.P. Cheng, Yau-Leung Ho,
Chung-Wai Lau.
Neuropsychology Team: Tatia M.C. Lee, Zoe Y.S. Sun,
Ming-wai Tsang, Candy K.W. Lok, Angel T.Y. Ip,
Angela Mou, Matthew W.Y. Tsang, Joy M.T. Yip.
r 2012 Lippincott Williams & Wilkins