British Journal of Anaesthesia 107 (5): 749–56 (2011)
Advance Access publication 8 August 2011 . doi:10.1093/bja/aer235
Occurrence of and risk factors for electroencephalogram
burst suppression during propofol– remifentanil anaesthesia
G. Besch 1, N. Liu 2, E. Samain 1,3,4, C. Pericard 1, N. Boichut 1, M. Mercier 3,5, T. Chazot 2 and S. Pili-Floury 1,3,4*
1
Department of Anesthesiology and Intensive Care Medicine, CHU Besancon, F-25000 Besancon, France
Department of Anesthesiology, Foch Hospital, F-92151 Suresnes, France
3
University of Franche-Comte, F-25000 Besancon, France
4
EA 3920, INSERM IFR 133, F-25000 Besancon, France
5
Department of Biostatistics, EA 3181, INSERM IFR 133, F-25000 Besancon, France
2
* Corresponding author. E-mail: [email protected]
Editor’s key points
† Burst suppression (BS)
from the EEG occurs
during deep anaesthesia,
but also in cases of
metabolic or
haemodynamic brain
injury.
† This study retrospectively
studied the occurrence of
BS during bispectral
index-controlled propofol
and remifentanil
anaesthesia.
† BS is mainly observed in
elderly male patients or
in patients with a history
of coronary artery
disease.
† The mechanisms
underlying BS and the
potential consequences
for the patient’s
postoperative outcome
remain unsolved.
Background. Suppression ratio (SR) derived from bispectral index (BIS) monitoring is
correlated to EEG burst suppression. It may occur during deep anaesthesia, but also in
the case of metabolic or haemodynamic brain injury. The goal of the study was to
describe the occurrence of SR and to determine factors associated with SR during
propofol –remifentanil general anaesthesia maintenance.
Methods. We conducted a post hoc analysis of BIS recordings in consecutive patients
included in two multi-centre trials, undergoing non-cardiac surgery using a dual closedloop BIS controller allowing automated propofol–remifentanil administration. The
percentage of time spent with a BIS value between 40 and 60 (TBIS 40 – 60) was measured.
Two groups of patients were defined: the SR group, including patients with at least one
episode of SR value .10% lasting more than 1 min, and the control group. Factors
associated with SR were analysed using a stepwise multivariate analysis.
Results. A total of 1494 patients [age¼57 (17) yr; TBIS 40 – 60 ¼76 (17%)] were analysed and
131 (8.7%) patients constituted the SR group. The main independent factors associated
with SR were advanced age [odds ratio (95% confidence interval)¼4.80 (1.85 –12.43)
(P¼0.027), 10.59 (3.76 –29.81) (P,0.0001), for categories of age 60 –80 and .80 yr,
respectively], history of coronary artery disease (CAD) [2.53 (1.47 –4.37) (P¼0.001)] and
male gender [1.57 (1.03 –2.40) (P¼0.03)].
Conclusions. The occurrence of SR during BIS-controlled propofol and remifentanil
anaesthesia is mainly observed in elderly male patients or in patients with a history of
CAD. The mechanisms underlying SR and the potential consequences for the patient’s
postoperative outcome remain unsolved.
Keywords: anaesthetic techniques, i.v. infusion; monitoring, depth of anaesthesia;
physiology, neurophysiology; potency, anaesthetic, age factors
Accepted for publication: 18 June 2011
The algorithm behind the bispectral index (BIS) value incorporates three sub-parameters: the relative b ratio, derived
from the power EEG spectrum; the SyncFastSlow, expressing
relative synchrony of fast and slow waves derived from bispectral analysis; and a suppression ratio (SR) that quantifies
the proportion of burst suppression (BS) EEG pattern or isoelectric activity.1 2 Several studies have reported a correlation
between too deep anaesthesia and poor outcome after
general anaesthesia. Monk and colleagues3 suggested that
a cumulative deep hypnotic time, defined by a BIS value
below 45, was an independent predictor of 1 yr mortality
after major non-cardiac surgery. Three recent studies also
strongly suggested an association between a cumulative
duration of intraoperative low BIS value and postoperative
mortality after cardiac4 5 or non-cardiac surgery.5 6
However, none of these studies reported the occurrence
of SR.
The BS EEG pattern is characterized by alternating periods
of normal to high voltage activity changing to low voltage
or even isoelectricity rendering the EEG inactive in appearance.7 8 The burst SR is a time domain EEG parameter developed to quantify this phenomenon.8 To calculate this
parameter, suppression is recognized as periods longer
than 0.50 s during which the EEG voltage does not exceed
approximately +5.0 mV. The time in a suppressed state is
measured, and the SR is reported as the fraction of the
& The Author [2011]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved.
For Permissions, please email: [email protected]
BJA
epoch length where the EEG is suppressed.9 Correlation
between the SR derived from BIS analysis and the BS
surface EEG pattern during general anaesthesia has been
reported.10 11 SR is believed to be a strong, synchronized
outflow of thalamic discharges to a widely unresponsive
cortex.12 The proportion of isoelectric periods is higher with
increasing anaesthetic drug concentrations, until eventually
the EEG pattern becomes completely isoelectric.13 14 This
suggests that SR could be an indicator of too deep anaesthesia.1 14 15 SR is also seen during coma or brain death but
never during physiological sleep.16 Several case reports also
suggest that SR could be associated with intraoperative
metabolic disorders such as hypothermia,17 18 hypoxia,19
hypoglycaemia or vascular brain injury,18 20 – 22 and to brain
death.23 – 26 However, little is known about the prevalence
and risk factors associated with the occurrence of SR
during BIS monitored anaesthesia.7 22 27 The main goal of
this study was to describe the onset, occurrence, and risk
factors for SR during general propofol–remifentanil anaesthesia maintenance.
Methods
Study design and BIS monitoring
We conducted a post hoc analysis of two European multi-centre
trials recording intraoperative BIS in 1494 patients undergoing
elective surgery (intracranial excepted) under general anaesthesia. Adult patients with cerebral disease or with chronic
use of psychotropic drugs, cardiac pacemaker patients and
pregnant or breast-feeding women were not included. Patients
(n¼914) were included in the PosoAnes trial (registered with
ClinicalTrials.gov, number NCT-00896714) and Boucle-N2O
trial (n¼580; NCT-00547209). For both trials, a dual closed-loop
controller was used allowing the automated titration of propofol and remifentanil guided by the BIS. The controller measures
and calculates the difference between the setpoint (BIS¼50)
and the measured BIS. If different from 0, the controller determines a new concentration of propofol, remifentanil or both
using the pharmacokinetic model of Schnider and colleagues28
for propofol and the model of Minto and colleagues29 for remifentanil. The error size determines which drug will be modified:
if the BIS error is small only the remifentanil is changed, if the
BIS error is higher, the two drug concentrations are modified.2
The PosoAnes trial was an observational study, designed to
identify factors influencing anaesthetic drug requirements
such as the time, type, and duration of surgery. We planned
to recruit 4500 patients. The Boucle-N2O trial was a randomized
controlled study designed to determine the sparing effect of
60% nitrous oxide on propofol and remifentanil requirements.
This trial demonstrated that the sparing effect of nitrous oxide
was limited to 6% for propofol. The occurrence of at least one
episode of SR.10% was not significantly different between
patients receiving nitrous oxide or not.2
In all patients, patient characteristic data, past medical
history, and duration of surgery were recorded. Upon arrival
in the operating theatre, a dedicated indwelling venous catheter was connected via a three-way Smartsitew (Alaris
750
Besch et al.
Medical Systems, San Diego, CA, USA), with a priming
volume 0.3 ml) to an infusion pump, and routine monitoring
was started. The BIS electrode (Zipprep, Covidien, Mansfield,
MA, USA) was positioned on the patient’s forehead and connected to either an A-2000 XP (version 3.11) BIS monitor
(Covidien) or a BIS Module (version 4.0 XP, GE-Healthcare,
Helsinki, Finland). For all patients, anaesthesia was performed using a dual closed-loop controller allowing automated propofol– remifentanil titration guided by the BIS.
A standard personal computer was used to provide a user
interface and to control communication with the BIS
monitor and with both propofol and remifentanil infusion
pumps (Alaris Medical, Hampshire, UK) via an RS232 serial
port (Infusion Toolbox 95w version 4.11 software).30 Electrode impedance was checked and the BIS sampling rate
was 256 Hz, with a 15 s smoothing rate. A valid BIS measurement was assumed when the signal quality index was .50.
Titration of anaesthetics was adjusted by the controller to
maintain a BIS value as close as possible to 50, and
between 40 and 60 (target range). The BIS-derived SR parameter was recorded.
Data collection and study groups
Every 5 s from the induction of anaesthesia to the end of
surgery, Signal Quality Index, BIS, SR values computed by
the BIS monitor, and propofol and remifentanil calculated
effect-site concentrations were automatically recorded on a
hard disk, and stored for subsequent analysis.
The following parameters were calculated: (i) TBIS 40 – 60 ¼
percentage of time spent in the BIS target range during
anaesthesia, defined as the ratio of the duration of anaesthesia with a BIS between 40 and 60 to total duration of
anaesthesia; (ii) TBIS.60 ¼percentage of anaesthesia time
spent with a BIS value above 60; and (iii) TBIS,40 ¼percentage
of anaesthesia time spent with a BIS value below 40.
Two groups of patients were defined a priori when designing the study: the SR group included the patients who had at
least one episode of SR value .10% lasting more than 1 min
consecutively during maintenance of anaesthesia, and the
control group included the remaining patients. For each
patient, the highest SR value was also identified, and we calculated the proportion of patients in each of the following SR
categories: 0–5, 6–10, 11– 15, 16–20, 21 –30, 31–40, and
.40. The normalized doses of propofol (mg kg21 h21) and
remifentanil (mg kg21 min21) infused were calculated.27
Analysis of BIS – SR relationship
For further insight into the relationship between BIS and SR,
we analysed BIS values and effect-site concentrations of propofol and remifentanil in the 5 min before, during (t0), and 1
min after SR occurrence (t+1).
Statistical analysis
Statistical analysis was performed in our Biostatistics Department using SAS 9.02 software (SAS Institute, Cary, NC, USA).
Results are expressed as mean (SD), median (inter-quartile
BJA
EEG burst suppression during i.v. anaesthesia
range), or number (percentage) as appropriate. Patient
characteristic data, duration of surgery, and percentage of
time spent in BIS target recorded in patients included in
the PosoAnes and Boucle-N2O trials were compared using
x 2 or Student’s t-tests as appropriate. The risk factors associated with the SR vs control groups were determined using
univariate analysis (x 2 or Student’s t-tests as appropriate).
The following factors were computed: age, gender, obesity
(defined as a BMI .30 kg m22), ASA physical status, preexisting co-morbidity, duration of surgery, average normalized dose of propofol and remifentanil, intraoperative use
of vasoactive drugs, and time spent in each BIS category. A
stepwise logistic regression was used to produce a risk
model for occurrence of SR and the Hosmer–Lemeshow statistic was used to measure calibration. Variables with a
P-value of ,0.2 in the univariate analysis were introduced
in the multivariate model. In the case of incomplete data,
patients were excluded from the multivariate analysis. To
estimate the potential risk of bias of selection, these patients
were compared with those included in the analysis for the
available data. For regression analysis, ASA status was
dichotomized (ASA I –II vs III –IV), age was recorded in
four categories (,40, 40–59, 60– 80, and .80 yr), and duration of surgery was recorded in three categories (≤2, 2– 4,
and .4 h). All P-values are two-sided and P,0.05 was considered statistically significant.
Results
Age distribution, duration of surgery, proportion of patients
with a past medical history of coronary artery disease
(CAD), and the average normalized dose of propofol of
patients included in PosoAnes or Boucle-N2O trials differ
only slightly (Table 1). Therefore, for the analysis, the data
of the 1494 patients [age¼57 (17) yr] could be aggregated.
We recorded 3742 h of closed-loop anaesthesia during
which 217 074 target modifications of propofol or remifentanil were made by the controller.
One hundred and thirty-one (8.7%) patients experienced
at least one episode of SR values .10% for more than
1 min during maintenance of anaesthesia, and constituted
the SR group. In these patients, a total of 257 episodes of
SR were recorded. The percentages of patients in each SR
category in the total number of patients and in the SR
group are given in Figure 1.
The result of the univariate analysis comparing the SR and
control groups is given in Table 2. Age, male gender, ASA
physical status, past medical history of CAD, congestive
heart failure or diabetes mellitus, TBIS 40 – 60, intraoperative
use of antihypertensive medication, and average doses of
anaesthetic were the variables included in the multivariate
analysis. The main independent factors associated with
occurrence of SR were advanced age, past medical history
of CAD, and male gender (Table 3). The distribution of different age categories for the SR group is shown in Figure 2.
The analysis of the individual recordings focused on the
period surrounding the occurrence of each episode of SR
allowed us to individualize three distinct patterns (Fig. 3A – C
and Table 4). Pattern A, observed in 118 (46%) cases, was
characterized by the recording of at least one BIS value
.60 in the 5 min period before SR occurrence (representative
recording shown in Fig. 3A). In pattern B, observed in 90 cases
(35%, Fig. 3C), BIS value was ,40 at SR occurrence (t0), and
no BIS value .60 was reached in the 5 min period of recording before SR occurrence. In pattern C, observed in 49 (19%)
cases, BIS values recorded in the 5 min period before occurrence and at onset (t0) of SR value .10 were within the
target range (Fig. 3B). The effect-site concentrations of propofol and remifentanil, calculated by the controller, and
their changes in the 5 min period were compared between
the three patterns using Tukey’s post hoc analysis and are
given in Table 4. Finally, 118 or 0.5‰ target modifications
by the closed-loop controller were followed by SR.
Discussion
The main finding of this study was that the occurrence of
transient periods of BS or isoelectric activity, attested by
the recording of SR value .10 for more than 1 min, was
observed in 8.7% of patients undergoing different types of
non-cardiac surgery during propofol –remifentanil anaesthesia. Main independent factors associated with SR were
advanced age, past medical history of CAD, and male
gender.
Ageing was an independent risk factor of SR, with an odds
ratio .10 in patients over 80 yr of age. This may be related to
the alteration in anaesthetic drug pharmacokinetic and
pharmacodynamic responses in the elderly.28 29 Interestingly, age-related SR occurred, although the dual closed-loop
controller that we used took into account the pharmacokinetic particularities of the older patients by using the
models of Minto and colleagues29 31 and Schnider and colleagues28 32 in order to reduce the risk of drug overdosing.
However, these two models were determined using a
limited number of healthy volunteers in clinically controlled
conditions and not in elderly patients suffering from several
co-morbidities. An alternative explanation of the age-related
occurrence of SR is linked to the modification of skull conductivity induced by ageing.33 Elderly patients could have
reduced skull conductivity that alters the EEG signal collection which could lead to a higher incidence of SR during
anaesthesia. On the other hand, our results show that SR
was less prevalent in women. Several studies have already
shown sex-related differences in the pharmacodynamic
effect of anaesthetic agents and support our findings.34 35
SR is a processed EEG marker of BS that is automatically
computed and displayed during BIS monitoring.10 SR is an
estimate of the suppressed EEG amount observed over the
last minute. The instantaneous BIS value is a weighted, nonlinear combination of core features, including the SR.8 The
weight of the SR to calculate BIS is itself a function of SR.
SR has no contribution to BIS for SR,10%, it has linearly
increasing weight as it increases, and completely determines
BIS when .50%.1 Using the 15 s smoothing rate, the EEG
751
BJA
Besch et al.
Table 1 Patient characteristics and anaesthetic procedure. Data are mean (SD), median (inter-quartile range). *Data are number (%). BIS,
bispectral index; SR, suppression ratio. TBIS,40, TBIS 40 – 60, TBIS.60, percentage of anaesthesia time spent with a BIS value below 40, between 40
and 60, and above 60, respectively
Characteristics
All patients
(n51494)
PosoAnes trial
(n5914)
Boucle-N2O trial
(n5580)
,40 yr
241 (16)
161 (18)
80 (14)
40 –59 yr
569 (38)
311 (34)
258 (45)
60 –80 yr
571 (38)
366 (40)
205 (35)
.80 yr
113 (8)
76 (8)
37 (6)
Male gender*
767 (51)
468 (51)
299 (52)
0.90
ASA physical status III–IV*
295 (20)
180 (20)
115 (20)
0.95
Coronary artery disease
91 (6)
42 (5)
49 (8)
0.002
Congestive heart failure
58 (4)
31 (4)
27 (5)
0.22
Chronic renal failure
90 (6)
57 (6)
33 (6)
0.67
Hypertension
405 (27)
240 (26)
165 (28)
0.35
Diabetes mellitus
106 (7)
65 (7)
41 (7)
0.50
BMI .30 kg m22
246 (16)
149 (16)
97 (17)
Age*
P-value
0.0006
Past medical history*
Duration of surgery*
0.84
0.0004
,2 h
711 (48)
470 (51)
241 (42)
2– 4 h
559 (37)
325 (36)
234 (40)
.4 h
224 (15)
119 (13)
105 (18)
TBIS,40
19 (17), 14 (8 –25)
16 (16), 7 (12 –21)
23 (17), 19 (11 –30)
,0.0001
TBIS 40 – 60
76 (17), 80 (67 – 88)
78 (17), 82 (71 – 90)
73 (18), 77 (64 –85)
,0.0001
TBIS.60
5 (7), 3 (1 –6)
Remifentanil dose (mg kg21 min21)
Propofol dose (mg kg21 h21)
SR group*
0.18 (0.08)
0.93
5.0 (1.7)
5.2 (1.8)
4.7 (1.5)
,0.0001
56 (6)
75 (13)
,0.0001
SR group
All patients
% of patients
60
50
40
30
20
10
<5
5–10 10–15 15–20 20–30 30–40
SR categories (%)
>40
Fig 1 Percentage of patients in each SR category.
used in calculating the current non-suppression component
of BIS uses the last 15 s of artifact-free EEG available in
the 1 min history buffer. So, if SR ,10% and assuming no
artifact in the recent past, BIS provides an average picture
of the patient state that is on average 7.5 s old (i.e. half
of the smoothing rate duration). If there is an artifact, then
752
,0.0001
0.18 (0.08)
80
0
4 (5), 3 (2 –5)
0.18 (0.08)
131 (9)
70
6 (8), 4 (2 –7)
the average age of the data used is older because the algorithm has to look at older data to find artifact-free data to
use. If there is more than 45 s of artifact in the last
minute, then there are insufficient data to calculate a BIS
index. If SR.50% so that BIS is completely determined by
SR, then the calculation window is roughly 1 min long. In
the absence of artifact and using a 15 s smoothing rate,
the average delay of EEG used in the BIS increases from
7.5 s for SR,10 to 30 s for SR.50. SR could easily be used
intraoperatively by anaesthetists without technical expertise
in EEG monitoring. However, intraoperative incidence of SR is
poorly described, as in most trials studying depth of anaesthesia, analysis is centred on BIS rather than on SR values.
Furthermore, its incidence may vary depending on the definition used.36 Thus, the impact of SR occurrence and of the
duration of time spent with SR is unknown. In the current
study, we chose to compare patients with and without at
least one value of SR.10% during more than 1 min, and
we acknowledge that this choice may be considered as arbitrary. It was made: (i) to avoid the inclusion of patients with a
very short period of SR, (ii) to limit the risk of including
patients with a single value of SR due to artifacts, and (iii)
because an SR value of .10 is the lowest value of SR computed in the BIS index calculation.
BJA
EEG burst suppression during i.v. anaesthesia
Table 2 Comparison of patient characteristic data and anaesthetic procedure between the SR and control groups. Data are mean (SD), median
(inter-quartile range). *Data are number (%). BIS, bispectral index; SR, suppression ratio. TBIS,40, TBIS 40 – 60, TBIS.60, percentage of anaesthesia
time spent with a BIS value below 40, between 40 and 60, and above 60, respectively
SR group (n5131)
Control group (n51363)
P-value
Age*
,0.0001
,40 yr
5 (4)
236 (17)
40 –59 yr
32 (24)
537 (39)
60 –80 yr
69 (53)
501 (37)
.80 yr
25 (19)
89 (7)
Male gender*
78 (59)
687 (50)
0.03
ASA physical status III–IV*
45 (34)
251 (18)
,0.0001
Coronary artery disease
29 (22)
62 (5)
,0.0001
Congestive heart failure
16 (12)
42 (3)
,0.0001
Past medical history*
Chronic renal failure
16 (12)
74 (5)
0.002
Hypertension
71 (53)
334 (25)
,0.0001
Diabetes mellitus
22 (17)
84 (8)
0.001
BMI .30 kg m22
21 (16)
225 (17)
0.82
,2 h
58 (44)
653 (48)
2– 4 h
52 (40)
506 (37)
.4 h
21 (16)
204 (15)
Duration of surgery*
0.73
TBIS,40
27 (20), 21 (13 –38)
18 (16), 14 (8– 24)
,0.0001
TBIS 40 – 60
65 (21), 71 (50 –81)
77 (17), 81 (69 –89)
,0.0001
TBIS.60
7 (11), 4 (2 –8)
Antihypertensive therapy*
31 (23)
59 (44)
Remifentanil dose (mg kg21 min21)
Propofol dose (mg kg21 h21)
0.16 (0.07), 0.15 (0.11 – 0.20)
4.8 (1.4), 4.6 (3.8 –5.4)
Table 3 Independent risk factors of intraoperative SR. Adjusted
on: ASA physical status (I –II vs III –IV), diabetes mellitus,
hypertension, chronic renal failure, congestive heart failure,
peroperative antihypertensive medication, peroperative
vasopressive medication, trial of inclusion, and propofol and
remifentanil dose (mg kg21 h21 and mg kg21 min21, respectively).
P¼0.67 for the Hosmer –Lemeshow goodness of fit test. TBIS 40 – 60,
per cent of time spent in the BIS target range between 40 and 60
Odds ratio (95% CI)
P-value
Age
1.00
40 –59 yr
2.16 (0.81 –5.75)
0.068
60 –80 yr
4.80 (1.85 –12.43)
0.027
10.59 (3.76 –29.81)
,0.0001
0.97 (0.96 –0.98)
,0.0001
TBIS 40 – 60
,0.0001
518 (38)
0.15
0.19 (0.08), 0.17 (0.13 – 0.23)
0.0004
5.0 (1.7), 4.7 (3.8 –6.0)
0.045
25
20
15
10
5
,40 yr
.80 yr
0.0003
122 (9)
% of patients
Ephedrine bolus*
Risk factors
5 (7), 3 (2– 6)
Coronary artery disease
No
1.00
Yes
2.53 (1.47 –4.37)
0
< 40
40–59
60–80
> 80
Age categories
Fig 2 Proportion of patients from the SR group in each category
of age.
0.001
Gender
Female
1.00
Male
1.57 (1.03 –2.40)
0.03
We found that SR.10 was frequently associated with a
deep level of anaesthesia attested by a BIS value of ,40
at the onset of SR (Table 4; patterns A and B).7 37 We found
753
BJA
Besch et al.
A
B
C
100
100
80
80
80
60
60
60
40
40
40
20
20
20
BIS and SR values
100
Effect-site concentration
Propofol (mg ml–1)
Remifentanil (ng ml–1)
0
16
14
12
10
8
6
4
2
0
85
85
95 102 105
t–5 min t0 t+1 min
95 102 105
t–5 min t0 t+1 min
115
0
75
t–5 min
115
8485
t0 t+1 min
95
SR
BIS
0
370
380 386 390
t–5 min t0 t+1 min
400
16
14
12
10
8
6
4
2
0
16
14
12
10
8
6
4
2
0
75
t–5 min
8485
t0 t+1 min
95
Anaesthetic time (min)
SR episode time
Remifentanil
Propofol
370
380 386 390
t–5 min t0 t+1 min
400
Anaesthetic time (min)
SR episode time
Fig 3 Individual recordings focused on the period surrounding the onset of each episode of SR value .10%. (A) Representative recording of
pattern A, characterized by at least one BIS value .60 in the 5 min period before SR occurrence; (B) representative of pattern B, characterized
by a BIS ,40 at SR occurrence (t0), and no BIS value .60 reached in the 5 min period of recording before SR occurrence; (C) representative of
pattern C, characterized by BIS values within target range in the 5 min period before, and at SR occurrence.
Table 4 BIS values and effect-site concentrations of propofol and
remifentanil in the three patterns of recordings individualized.
Data are median (inter-quartile range). Pattern A, the recording of
at least one BIS value .60 in the 5 min period before SR
occurrence. Pattern B, BIS,40 at SR occurrence (t0), and no BIS
value .60 reached in the 5 min period of recording before SR
occurrence. Pattern C, BIS values within target range in the 5 min
period before, and at SR occurrence. t0, time of SR occurrence; BIS,
bispectral index. *Pattern A vs pattern C (P,0.05); †pattern A vs
pattern B (P,0.05); ‡pattern B vs pattern C (P,0.05)
Pattern A
(n5118
episodes)
Pattern B
(n590
episodes)
Pattern C
(n549
episodes)
BIS value at t0
29 (25 – 34)
28 (24 – 33)
45 (41 –48)
BIS value at
t0+1 min
27 (23 – 32)
25 (22 – 31)
41 (33 –46)
Effect-site concentration at t0
Remifentanil
(ng ml21)
8.4 (5.5 –
12.0)*,†
6.0 (4.0 –
9.6)†,‡
12.0 (7.8 –
14.2)*,‡
Propofol
(mg ml21)
4.2 (2.7 –5.0)*
3.4 (2.1 –4.5)‡
4.5 (3.0– 5.0)*,‡
% of effect-site concentration change
Remifentanil
16 (25 to 52)
11 (24 to 33)
13 (0– 26)
Propofol
26 (0 –66)
25 (0 –78)
13 (0– 40)
that 46% of the SR episodes were associated with at least
one episode of BIS value above 60 in the 5 min period
before SR occurrence (pattern A), followed by a significant
increase in effect-site concentrations of anaesthetic agents.
The fact that SR may have been favoured by the rapid deepening of anaesthesia induced by the closed-loop controller in
754
the case of BIS value.60 or a rapid BIS increase could be
expected as these two factors are taken into consideration
by the algorithm of the system, whereas it does not
compute SR to adjust anaesthetic infusion rates.2 In some
cases, SR occurred when BIS values in the 5 previous
minutes were within the target range (pattern C), suggesting
that the anaesthesia controller should not be considered as
the sole mechanism of SR occurrence. Furthermore, during
the 3742 h of BIS recording, more than 210 000 changes in
effect-site concentration of propofol or remifentanil were
made by the controller without SR occurrence. As such,
other factors may be involved in SR occurrence in these
cases, and several hypotheses can be raised. Hypovolaemia
or changes in cardiac output have been shown to modify
the pharmacokinetic –pharmacodynamic models of the
anaesthetic drugs and may have favoured the occurrence
of both deep anaesthesia and SR.38 On the other hand,
haemodynamic impairment may induce a decrease in cerebral blood flow that may favour SR. In this regard, several
case reports or short series have shown a relationship
between SR and cerebral injury from various origins, including traumatic injury or hypoxaemia.19 20 39 40 Our study
was not designed to address this issue, as intraoperative
haemodynamic parameters or patient outcome was not
recorded. SR monitoring could also identify periods of excessive anaesthesia, not detected by the calculation of the BIS
value itself. One hypothesis is that propofol anaesthesia
specifically induced spindle activity that does not exist
during anaesthesia using volatile agents.41 42 The frequency
of the spindles is usually in the range of 13 –15 Hz,43 similar
to alpha or beta activity ranges. These spindles could artifact
the EEG activity, interpreted by the BIS monitor and
BJA
EEG burst suppression during i.v. anaesthesia
artificially increase the BIS value measured without any
effect on the SR measurement. Another hypothesis is supported by the fact that most of the patients concerned in
this discrepancy showed a high effect-site concentration of
remifentanil (Table 4 and Fig. 3C). A specific effect of such a
concentration of remifentanil on either propofol concentration, recently reported by Yufune and colleagues, or on
EEG signal, as suggested by other studies, could lead to SR
without significant impact on the BIS value.44 45 However,
interaction between the opioid and hypnotic was not specifically addressed in this study.
As this is a post hoc analysis of previous clinical trials, this
study carries some limitations. Patients were primarily included
in two different prospective, multi-centre European trials.
Nitrous oxide was used in patients in the Boucle-N2O trial.
BIS40 – 60, BIS,40, or the occurrence of SR were similar between
the two groups despite the use of nitrous oxide. We consider
that we can aggregate patients included in these two trials
since the same inclusion and exclusion criteria were used.
Another limitation is that we arbitrarily chose an SR
threshold of 10%, although we do not know if such an SR
value could be harmful. However, the effect of ageing on
the presence of SR was unchanged whatever the SR
threshold chosen. Anaesthesia was conducted using a dual
closed-loop of propofol and remifentanil in all patients that
permitted maintenance of anaesthesia in the target range
in 76% of total anaesthetic time. The performance of the
controller can be considered as adequate if compared with
manual control of anaesthetic infusion by anaesthetists,
reported in the literature, with a percentage of time spent
in the target ranging from 49% to 77%.46 – 49 Moreover, we
recently reported that manual control of propofol and remifentanil increases by three-fold the occurrence of SR when
compared with the closed-loop control.2 On the other
hand, these results might not be applicable to patients
receiving volatile anaesthesia. We have not recorded the
raw EEG, and thus cannot eliminate that the occurrence of
SR was related to artifact, decrease in EEG total power in
an ageing population, or spindles. Studies including simultaneously raw EEG, continuous arterial pressure, intra-cranial
perfusion, and brain metabolism are necessary to determine
the weight of these different parameters in SR genesis.
Patient outcome was not recorded.
In conclusion, during BIS-controlled propofol and remifentanil anaesthesia, the occurrence of SR value .10 during
more than 1 min is mainly observed in elderly male patients.
The presence of SR could be an early sign of too deep anaesthesia. The question of the mechanisms of these periods of
EEG BS or isoelectric activity is unknown and can be related
to haemodynamics, metabolism, or drug overdosing. The
potential consequences for the patient’s postoperative
outcome should be established in a prospective study.
Acknowledgements
We thank warmly Alexandre Mebazaa for his help in preparing the manuscript.
Conflict of interest
N.L. and T.C. are patent holders in France for the gain constants and the control algorithm (No. BFF80P669, Institut
National de la Propriété Industrielle, France).
Funding
The work was not supported by official funding or a grant.
The financial cost of the study was supported by the Department of Anesthesiology and Intensive Care Medicine, CHU
Besancon, Besancon, France.
References
1 Bruhn J, Bouillon TW, Shafer SL. Bispectral index (BIS) and burst
suppression: revealing a part of the BIS algorithm. J Clin Monit
Comput 2000; 16: 593–6
2 Liu N, Chazot T, Hamada S, et al. Closed-loop coadministration of
propofol and remifentanil guided by Bispectral Index: a randomized multicenter study. Anesth Analg 2011; 112: 546–57
3 Monk TG, Saini V, Weldon BC, Sigl JC. Anesthetic management
and one-year mortality after noncardiac surgery. Anesth Analg
2005; 100: 4– 10
4 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– 27
5 Leslie K, Myles PS, Forbes A, Chan MT. The effect of bispectral
index monitoring on long-term survival in the B-aware trial.
Anesth Analg 2009; 110: 816– 22
6 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– 12
7 Morimoto Y, Hagihira S, Koizumi Y, Ishida K, Matsumoto M,
Sakabe T. The relationship between bispectral index and electroencephalographic parameters during isoflurane anesthesia.
Anesth Analg 2004; 98: 1336– 40
8 Rampil IJ, Kim JS, Lenhardt R, Negishi C, Sessler DI. Bispectral EEG
index during nitrous oxide administration. Anesthesiology 1998;
89: 671– 7
9 Rampil IJ. A primer for EEG signal processing in anesthesia.
Anesthesiology 1998; 89: 980– 1002
10 Riker RR, Fraser GL, Wilkins ML. Comparing the bispectral
index and suppression ratio with burst suppression of
the electroencephalogram during pentobarbital infusions in
adult intensive care patients. Pharmacotherapy 2003; 23:
1087– 93
11 Cottenceau V, Petit L, Masson F, et al. The use of bispectral index
to monitor barbiturate coma in severely brain-injured patients
with refractory intracranial hypertension. Anesth Analg 2008;
107: 1676– 82
12 Steriade M, Amzica F, Contreras D. Cortical and thalamic cellular
correlates of electroencephalographic burst-suppression. Electroencephalogr Clin Neurophysiol 1994; 90: 1– 16
13 Koitabashi T, Ouchi T, Umemura N. The effect of nitrous oxide on
the central nervous system evaluated by the bispectral index
under various levels of propofol anesthesia. Masui 2004; 53:
650– 3
755
BJA
14 Yoshitani K, Kawaguchi M, Takahashi M, Kitaguchi K, Furuya H.
Plasma propofol concentration and EEG burst suppression ratio
during normothermic cardiopulmonary bypass. Br J Anaesth
2003; 90: 122– 6
15 Hartikainen KM, Rorarius M, Perakyla JJ, Laippala PJ, Jantti V. Cortical reactivity during isoflurane burst-suppression anesthesia.
Anesth Analg 1995; 81: 1223–8
16 Brown EN, Lydic R, Schiff ND. General anesthesia, sleep, and
coma. N Engl J Med 2010; 363: 2638–50
17 Hayashida M, Sekiyama H, Orii R, et al. Effects of deep hypothermic circulatory arrest with retrograde cerebral perfusion on electroencephalographic bispectral index and suppression ratio.
J Cardiothorac Vasc Anesth 2007; 21: 61 –7
18 Seder DB, Fraser GL, Robbins T, Libby L, Riker RR. The bispectral
index and suppression ratio are very early predictors of neurological outcome during therapeutic hypothermia after cardiac arrest.
Intensive Care Med 2010; 36: 281–8
19 Liu N, Chazot T, Mutter C, Fischler M. Elevated burst suppression
ratio: the possible role of hypoxemia. Anesth Analg 2006; 103:
1609– 10
20 Declerck A, Liu N, Chazot T, Fischler M. BIS values during resuscitation: the role of the suppression ratio (case report). J Clin Monit
Comput 2009; 23: 307– 9
21 Johansen JW. Esmolol promotes electroencephalographic burst
suppression during propofol/alfentanil anesthesia. Anesth Analg
2001; 93: 1526– 31
22 Umegaki N, Hirota K, Kitayama M, Yatsu Y, Ishihara H, Mtasuki A.
A marked decrease in bispectral index with elevation of suppression ratio by cervical haematoma reducing cerebral perfusion
pressure. J Clin Neurosci 2003; 10: 694– 6
23 Vivien B, Langeron O, Riou B. Entropy and bispectral index
in brain-dead organ donors. Intensive Care Med 2007; 33:
919– 20
24 Vivien B, Paqueron X, Le Cosquer P, Langeron O, Coriat P,
Riou B. Detection of brain death onset using the bispectral
index in severely comatose patients. Intensive Care Med
2002; 28: 419– 25
25 Escudero D, Otero J, Muniz G, et al. The Bispectral Index scale: its
use in the detection of brain death. Transplant Proc 2005; 37:
3661– 3
26 Misis M, Raxach JG, Molto HP, Vega SM, Rico PS. Bispectral index
monitoring for early detection of brain death. Transplant Proc
2008; 40: 1279– 81
27 Liu N, Chazot T, Genty A, et al. Titration of propofol for anesthetic
induction and maintenance guided by the bispectral index:
closed-loop versus manual control: a prospective, randomized,
multicenter study. Anesthesiology 2006; 104: 686–95
28 Schnider TW, Minto CF, Shafer SL, et al. The influence of age
on propofol pharmacodynamics. Anesthesiology 1999; 90:
1502– 16
29 Minto CF, Schnider TW, Egan TD, et al. Influence of age and gender
on the pharmacokinetics and pharmacodynamics of remifentanil.
I. Model development. Anesthesiology 1997; 86: 10– 23
30 Cantraine FR, Coussaert EJ. The first object oriented monitor
for intravenous anesthesia. J Clin Monit Comput 2000; 16:
3–10
31 Minto CF, Schnider TW, Shafer SL. Pharmacokinetics and pharmacodynamics of remifentanil. II. Model application. Anesthesiology
1997; 86: 24–33
756
Besch et al.
32 Schnider TW, Minto CF, Gambus PL, et al. The influence of
method of administration and covariates on the pharmacokinetics of propofol in adult volunteers. Anesthesiology 1998;
88: 1170–82
33 Wendel K, Vaisanen J, Seemann G, Hyttinen J, Malmivuo J. The
influence of age and skull conductivity on surface and subdermal
bipolar EEG leads. Comput Intell Neurosci 2010: 1– 7, ID 397272,
doi: 10.1155/2010/397272
34 Gan TJ, Glass PS, Sigl J, et al. Women emerge from general
anesthesia with propofol/alfentanil/nitrous oxide faster than
men. Anesthesiology 1999; 90: 1283– 7
35 Hoymork SC, Raeder J. Why do women wake up faster than men
from propofol anaesthesia? Br J Anaesth 2005; 95: 627–33
36 Soehle M, Kuech M, Grube M, et al. Patient state index vs bispectral index as measures of the electroencephalographic effects of
propofol. Br J Anaesth 2010; 105: 172– 8
37 Hartikainen K, Rorarius M, Makela K, Yli-Hankala A, Jantti V. Propofol and isoflurane induced EEG burst suppression patterns in
rabbits. Acta Anaesthesiol Scand 1995; 39: 814–8
38 Takizawa E, Takizawa D, Hiraoka H, Saito S, Goto F. Disposition and
pharmacodynamics of propofol during isovolaemic haemorrhage
followed by crystalloid resuscitation in humans. Br J Clin Pharmacol 2006; 61: 256–61
39 Myles PS, Cairo S. Artifact in the bispectral index in a patient with
severe ischemic brain injury. Anesth Analg 2004; 98: 706–7
40 Savard M, Huot P. Asynchronous burst-suppression on EEG in
severe, diffuse axonal injury. Can J Neurol Sci 2008; 35: 526– 7
41 Ferenets R, Lipping T, Suominen P, et al. Comparison of the properties of EEG spindles in sleep and propofol anesthesia. Conf Proc
IEEE Eng Med Biol Soc 2006; 1: 6356–9
42 Jantti V, Yli-Hankala A, Baer GA, Porkkala T. Slow potentials of EEG
burst suppression pattern during anaesthesia. Acta Anaesthesiol
Scand 1993; 37: 121–3
43 Wolter S, Friedel C, Bohler K, Hartmann U, Kox WJ, Hensel M. Presence of 14 Hz spindle oscillations in the human EEG during deep
anesthesia. Clin Neurophysiol 2006; 117: 157–68
44 Kortelainen J, Koskinen M, Mustola S, Seppanen T. Effects of remifentanil on the spectrum and quantitative parameters of electroencephalogram in propofol anesthesia. Anesthesiology 2009;
111: 574–83
45 Yufune S, Takamatsu I, Masui K, Kazama T. Effect of remifentanil on
plasma propofol concentration and bispectral index during propofol anaesthesia. Br J Anaesth 2011; 106: 208–14
46 Phillips AS, McMurray TJ, Mirakhur RK, Gibson FM, Elliott P. Propofol– fentanyl anaesthesia in cardiac surgery: a comparison in
patients with good and impaired ventricular function. Anaesthesia 1993; 48: 661– 3
47 Puri GD, Kumar B, Aveek J. Closed-loop anaesthesia delivery
system (CLADS) using bispectral index: a performance assessment study. Anaesth Intensive Care 2007; 35: 357–62
48 Struys MM, De Smet T, Versichelen LF, Van De Velde S, Van den
Broecke R, Mortier EP. Comparison of closed-loop controlled
administration of propofol using Bispectral Index as the controlled variable versus ‘standard practice’ controlled administration. Anesthesiology 2001; 95: 6– 17
49 Leslie K, Sleigh J, Paech MJ, Voss L, Lim CW, Sleigh C. Dreaming
and electroencephalographic changes during anesthesia maintained with propofol or desflurane. Anesthesiology 2009; 111:
547– 55