Dexmedetomidine vs propofol-remifentanil conscious sedation for awake craniotomy: A prospective randomized controlled trial

Abstract

Background: Awake craniotomy (AC) is performed for the resection of brain tumours in close proximity to areas of eloquent brain function to maximize reduction of tumour mass and minimize neurological injury. This study compares the efficacy and safety of dexmedetomidine vs propofol-remifentanil-based conscious sedation, during AC for supratentorial tumour resection.
Methods: Prospective, randomized, controlled trial including 50 adult patients undergoing AC who were randomly assigned to a dexmedetomidine (DEX group, n=25) or propofol-remifentanil group (P-R group, n=25). The primary outcome was the ability to perform intraoperative brain mapping assessed on a numeric rating scale (NRS). Secondary outcome was the efficacy of sedation measured by the modified Observer's Assessment of Alertness/Sedation (OAA/S) scale. Other outcome measures including haemodynamic and respiratory variables, pain, sedation and anxiety scores, adverse events, and patient satisfaction were also compared.
Results: There were no differences between DEX and P-R groups regarding the ability to perform intraoperative brain mapping [mean NRS score (95% CI): 10.0 (9.9–10.0) vs 9.7 (9.5–10.0), P=0.13] and level of sedation during mapping [mean OAA/S score (95% CI): 4.1 (3.5–4.7) vs 4.3 (3.9–4.7), P=0.51], respectively. Respiratory adverse events were more frequent in the P-R group (20 vs 0%, P=0.021). Heart rate was significantly lower in the DEX group across time (P<0.001); however, the need for treatment of bradycardia was not different between groups.
Conclusions: Quality of intraoperative brain mapping and efficacy of sedation with dexmedetomidine were similar to propofol-remifentanil during AC for supratentorial tumour resection. Dexmedetomidine was associated with fewer respiratory adverse events.
Clinical trial registration: NCT01545297.

Full text

NEUROSCIENCES AND NEUROANAESTHESIA
Dexmedetomidine vs propofol-remifentanil conscious
sedation for awake craniotomy: a prospective
randomized controlled trial
†
‡
N. Goettel1,3,4, S. Bharadwaj1, L. Venkatraghavan1, J. Mehta1, M. Bernstein2
and P. H. Manninen1,*
1
Department of Anesthesia, Toronto Western Hospital,
2
Division of Neurosurgery, Toronto Western Hospital,
University Health Network, University of Toronto, Toronto, Canada,
3
Department of Anesthesia, Surgical
Intensive Care, Prehospital Emergency Medicine and Pain Therapy and
4
Department of Clinical Research,
University Hospital Basel, University of Basel, Basel, Switzerland
*Corresponding author. E-mail: pirjo.manninen@uhn.ca
Abstract
Background: Awake craniotomy (AC) is performed for the resection of brain tumours in close proximity to areas of eloquent
brain function to maximize reduction of tumour mass and minimize neurological injury. This study compares the efficacy and
safety of dexmedetomidine vs propofol-remifentanil-based conscious sedation, during AC for supratentorial tumour resection.
Methods: Prospective, randomized, controlled trial including 50 adult patients undergoing AC who were randomly assigned to a
dexmedetomidine (DEX group, n=25) orpropofol-remifentanil group(P-R group, n=25). The primary outcomewas the ability to perform
intraoperative brain mapping assessed on a numeric rating scale (NRS). Secondary outcomewas the efficacy of sedation measured by
the modified Observer’s Assessment of Alertness/Sedation (OAA/S) scale. Other outcome measures including haemodynamic and
respiratory variables, pain, sedation and anxiety scores, adverse events, and patient satisfaction were also compared.
Results: There were no differences between DEX and P-R groups regarding the ability to perform intraoperative brain mapping
[mean NRS score (95% CI): 10.0 (9.9–10.0) vs 9.7 (9.5–10.0), P=0.13] and level of sedation during mapping [mean OAA/S score (95%
CI): 4.1 (3.5–4.7) vs 4.3 (3.9–4.7), P=0.51], respectively. Respiratory adverse events were more frequent in the P-R group (20 vs 0%,
P=0.021). Heart rate was significantly lower in the DEX group across time (P<0.001); however, the need for treatment of
bradycardia was not different between groups.
Conclusions: Quality of intraoperative brain mapping and efficacy of sedation with dexmedetomidine were similar to propofol-
remifentanil during AC for supratentorial tumour resection. Dexmedetomidine was associated with fewer respiratory adverse events.
Clinical trial registration: NCT01545297.
Key words: anaesthetics, intravenous; conscious sedation; craniotomy; dexmedetomidine; propofol; remifentanil
†
Euroanaesthesia Congress, May 31, 2015, Berlin, Germany, and Canadian Anesthesiologists’Society Annual Meeting, June 20, 2015, Ottawa, Canada.
‡This Article is accompanied by Editorial Aew113.
Accepted: January 3, 2016
© The Author 2016. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved.
For Permissions, please email: journals.permissions@oup.com
British Journal of Anaesthesia, 116 (6): 811–21 (2016)
doi: 10.1093/bja/aew024
Advance Access Publication Date: 20 April 2016
Neurosciences and Neuroanaesthesia
811
at University of Basel / A280 UKBB on October 2, 2016http://bja.oxfordjournals.org/Downloaded from

Editor’s key points
•For brain tumours in close proximity to eloquent areas,
intraoperative mapping can help optimize outcomes.
•To facilitate this, an ‘awake craniotomy’technique is per-
formed to facilitate wakefulness during mapping.
•The optimal sedation or anaesthetic technique for awake
craniotomy has not been identified.
•In this randomized controlled trial the authors compared
dexmedetomidine and propofol-remifentanil techniques.
Awake craniotomy (AC) is an accepted procedure for resection of a
brain tumour, located in close proximity to areas of eloquent brain
function, to achieve maximal surgical reduction of tumour mass
without injuring important functional areas of the brain, such as
the motor, language, or sensory cortex.
1–4
A variety of anaesthetic
techniqueshave been used for AC, ranging from an ‘asleep-awake-
asleep’technique, with or without mechanical ventilation, to the
managementof ‘fully awake’patients with local or regionalanaes-
thesia of the scalp.
56
The required level of sedation and analgesia
varies throughout the different stages of surgery, but most im-
portantly, the patient needs to be awake and alert during brain
mapping.
7
Different i.v. sedative drugs have been used in AC; for
conscious sedation or monitored anaesthesia care, many anaes-
thetists choose a combination of propofol and an ultra-short-
acting opioid such as remifentanil.
8–11
However, in AC patients
with an unsecured airway, the use of propofol sedation in combin-
ation with opioids has been associated with intraoperative airway
and/or respiratory complications, and poor patient cooperation
during cortical mapping.
912–14
Dexmedetomidine is a potent, highlyselective α
2
-adrenoceptor
agonist
15–17
with sedative, anxiolytic, analgesic, opioid-sparing,
18
and sympatholytic effects.
16
In contrast to other sedative agents,
dexmedetomidine is not associated with respiratory depression.
16 19
As a result of predictable pharmacokinetics and a rapid distribu-
tion half-life of 5–6 min
15 17
after bolus injection, dexmedetomi-
dine may be titrated to a desired effect. Prolonged infusions of
dexmedetomidine, however, may lead to delayed sedative effects
after discontinuation of the drug because of a longer context-sen-
sitive half-life.
20–23
The hypnotic properties of dexmedetomidine
are mediated via hyperpolarization of noradrenergic neurons in
the locus ceruleus. Fundamental research suggests that dexmedeto-
midine converges on a natural sleep pathway to exert its sedative
effect.
24
This unique state of sedation, also called ‘collaborative
sedation’,
25
may be useful for AC, which requires a deep level of
sedation during painful and stimulating operative procedures on
the one hand, and sufficient patient cooperation during mapping
of eloquent function on the other.
The purpose of thisstudy wasto compare the use of dexmede-
tomidine vs propofol-remifentanil-based conscious sedation, in
patients undergoing AC for the resection of supratentorial brain
tumours. We hypothesized that there would be no difference in
the ability to perform intraoperative brain mapping between dex-
medetomidine and propofol-remifentanil, and that both sedation
techniques would have comparable efficacy and safety profiles.
Methods
Trial design
The University Health Network Research Ethics Board provided
ethical approval for this study (Ethical Committee No. 11-0607-
A). All study participants provided written informed consent.
We conducted a prospective, double-blind, randomized trial. It
was conducted according to the revised Declaration of Helsinki
of the World Medical Association and ICH GCP guidelines for
good clinical trial practice. The study was registered on Clinical-
Trials.gov (NCT01545297) before patient enrolment.
Participants and study setting
Study participants were recruited at the Toronto Western
Hospital, University Health Network, Toronto, Canada. We in-
cluded patients aged ≥18 yr, ASA physical status I–III, undergoing
elective AC for the resection of a supratentorial brain tumour,
using a conscious sedation technique. Exclusion criteria werese-
vere cardiovascular or respiratory disease (ASA grade ≥IV), preg-
nancy, allergies to the drugs being used, known alcohol or
substance abuse, and expected communication difficulties with
the patient.
Interventions
Before surgery, 50 eligible patients were equally randomized to
receive either dexmedetomidine (DEX group) or propofol-remi-
fentanil (P-R group) infusions. The loading dose of dexmedetomi-
dinewas1µgkg
−1
over 10 min, followed by a maintenance
infusion titrated to effect (doses ranging from 0.2–1µgkg
−1
h
−1
). Continuous infusion rates of propofol and remifentanil
were 25–150 and 0.01–0.1µgkg
−1
min
−1
, respectively. Dosing of
all study drugs for surgical stages other than brain mapping
was adjusted to achieve a targeted level of sedation of 2–4 points,
on the modified Observer’s Assessment of Alertness/Sedation
(OAA/S) scale.
26
Anaesthetic management
Intraoperative anaesthetic management was standardized by
using the predefined sedation protocols in both groups. No pre-
medication was used. The patient was comfortably positioned
(supine or lateral) on the operating table. Vital signs were re-
corded using ASA standard monitors: non-invasive bp monitor-
ing, ECG, and pulse oximetry (SpO
2
). Arterial lines or urinary
catheters were not inserted routinely. All patients were breathing
spontaneously and received supplemental oxygen at 4 l min
−1
(inducing a mean inspired fraction of oxygen of approximately
36%) via nasal prongs. Naso- or oropharyngeal airway devices
were not used. The presence of end-tidal carbon dioxide (EtCO
2
)
was monitored at the oxygen delivery nasal prongs port to deter-
mine respiratory rate (RR).
After establishment of peripheral venous access in the operat-
ing room, each patient received fentanyl 50 µg i.v., and then the
study drug infusions were started according to the respective
sedation protocol. Approximately 10 min later, the sites of pin in-
sertion for rigid head fixation (Sugita frame) were infiltrated with
local anaesthetic agent (2% lidocaine with 1:200,000 epinephrine)
by the neurosurgeon. Infiltration of the scalp was performed
using 0.25% bupivacaine with 1:200,000 epinephrine to produce
a‘ring block’around the incision. The overall management of
the anaesthetic with respect to adjustments of the drug infusions
and the administration of all other required medications was
left up to the attending anaesthetist. At any time during the
procedure, when excessive pain was expected, or if the patient
complained of pain or discomfort, the infusion rates of dexmede-
tomidine (DEX group) or remifentanil (P-R group) were increased.
If necessary, additional fentanyl 25–50 µg i.v. was administered.
If sedation was inadequate in either group, the infusion rates
were increased at first. Rescue medication consisting of a propo-
fol bolus (20–30 mg i.v.) was given when first-line treatment
812 |Goettel et al.
at University of Basel / A280 UKBB on October 2, 2016http://bja.oxfordjournals.org/Downloaded from

failed. Ten min before brain mapping, propofol wasdisconti nued,
and dexmedetomidine and remifentanil infusions were reduced.
Minimal infusion rates of dexmedetomidine (0.1–0.4 µg kg
−1
h
−1
)
in the DEX group, and remifentanil (0.01–0.05 µg kg
−1
min
−1
)in
the P-R group were continued during mapping. Mapping for
motor, sensory and/or speech functions was performed after
placement of a stimulating electrode on the cortical surface by
the neurosurgeon.
2
The anaesthetist observed for any move-
ments of the face, arm or leg. Motor strength was tested by asking
the patient to move their hand (fingers) or foot (dorsiflexion)
against resistance. Patients were advised to note any changes
in sensation. Language was tested by asking the patient to
count or name lists of objects while observing for speech arrest
or hesitation. The duration of brain mapping was approximately
10 min. Subsequently, study drug infusions were resumed for tu-
mour resection and closure of the craniotomy. Patients received
fentanyl 0.5–1µgkg
−1
i.v. if they complained of headache or other
pain at the end of the procedure.
After surgery, patients were monitored in the postanaesthetic
care unit (PACU) for 2 h before being discharged to the ward or day
surgery unit. In the PACU, all standard monitoring of a neuro-
logical patient was performed, and postoperative pain was trea-
ted according to a standard protocol with a combination of oral
acetaminophen and morphine or fentanyl i.v. or oral oxycodone.
Ondansetron 4 mg, and/or dimenhydrinate 50 mg, and/or meto-
clopramide 20 mg and/or dexamethasone 4 mg i.v. were adminis-
tered for postoperative nausea and vomiting when needed. After
discharge from the PACU, the care of the patientincluding the ad-
ministration of analgesics and discharge from the hospital was
determined by the surgical team.
Outcome variables
The primary outcome measure was the quality of intraoperative
brain mapping. The ability of the patient to cooperate and per-
form cortical mapping was assessed on a 10-point numerical rat-
ing scale (NRS; 0=unsatisfactory; 10=excellent). Mapping was
considered successful when the NRS score was ≥8. The level of
sedation was recorded at the time of mapping and throughout
the procedure using the modified OAA/S scale. Using visual ana-
logue scales (VAS), patients were asked to evaluate levels of pain
(0=no; 1–3=mild; 4–6=moderate; 7–10=severe pain) and anxiety
(0–1=no or mild; 2–3=moderate; 4–5=severe anxiety). This assess-
ment was repeated at 12 successive time points throughout the
procedure (T0, baseline; T1, headpin insertion; T2, 5 min after
T1; T3, local anaesthetic infiltration to incision; T4, skin incision;
T5, craniotomy (bone work); T6, dura opening; T7, brain mapping;
T8, start of tumour resection; T9, 30 min after T8; T10, skin clos-
ure; T11, admission to PACU; and T12, 120 min after T11).
Secondary outcome measures included the incidence of ad-
verse events such as respiratory depression or airway obstruc-
tion, haemodynamic instability, failure to provide adequate
analgesia, and all intra- and immediate postoperative complica-
tions. Heart rate (HR), mean arterial pressure (MAP), SpO2, and RR
were recorded at the 12 successive time points (T0–T12). Haemo-
dynamic instability (arterial hypertension or hypotension, car-
diac arrhythmia) and respiratory events (airway obstruction,
apnoea/hypoventilation, oxygen desaturation), were defined as
an adverse event when a treatment intervention (administration
of a pharmacological agent for haemodynamic events, airway
manoeuvres and/or diminution of study drug infusion for re-
spiratory events) was required.
Preoperative variables included basic patient characteristics,
clinical characteristics and medical co-morbidities. Assessment
of the condition of the brain (lax or tight) upon opening of the
dura mater and any intraoperative neurological complication
(e.g., seizures, or new onset neurological deficits) were noted.
Other intraoperative patient complaints or events (e.g., cold/
shivering, nausea and vomiting, restlessness, fatigue, and need
for conversion to general anaesthesia) were also recorded. In
the PACU, the amount of opioid and antiemetic administered
and the incidence of adverse events were noted. Testing of mem-
ory and cognitive function was also performed using the Short
Portable Mental Status Questionnaire (SPMSQ
27
; Supplementary
data, Table S1) at 2 and 24 h after surgery. At 24 h after surgery,
the patients were interviewed in person and asked regarding
any adverse events such as excessive pain, nausea and vomiting.
They were asked how satisfactory were their intraoperative pain
management and overall level of comfort, recall of the intrao-
perative experience including pain, anxiety and discomfort, and
their willingness to repeat surgery, if needed, using the same an-
aesthetic technique. If the patient had been discharged home the
day of surgery, a telephone interview was conducted. Length of
hospital stay and final postoperative destination of patients (in-
or outpatient surgery, need for unplanned postoperative hospital
admission) were noted.
Sample size
A change of 25% in the ability to perform satisfactory intraopera-
tive brain mapping was considered to be of clinical importance.
To detect a mean difference of 2.5 points on the 10-point NRS
for mapping quality between the DEX and P-R groups, a sample
size of 25 subjects per group was required (total of 50 subjects),
considering a 2-sided test with α=0.05, power of 90%, standard de-
viation of 1, and assuming a 10% drop-out rate.
Randomization
We performed simple randomization of participants to the DEX
and P-R groups. One investigator generated the random alloca-
tion sequence and provided allocation concealment by using
sequentially numbered, sealed, opaque envelopes. A second in-
vestigator implemented the randomization method and enrolled
participants.
Blinding
A blinded investigator that was not directly involved in the an-
aesthetic management of the patients, collected all intra- and
postoperative data. Patient and neurosurgeon were blinded to
group allocation; however, it was not practical to blind the at-
tending anaesthetist to preoperative and intraoperative data, as
this information was essential for the medical care of patients.
For blinding purposes, two drug infusion pumps were used in
every patient. Study drug infusion pumps and i.v. connection
lines were concealed to avoid identification.
Statistical analysis
Analysis was performed using SAS statistical software, version
9.3 (SAS Institute, Cary, NC, USA). All analyses were undertaken
on a modified intention-to-treat set, comprising all patients
who had a baseline value during the intraoperative assessment.
Continuous variables and univariate differences between DEX
and P-R groups were compared using the Wilcoxon rank-sum
test, categorical variables using the χ
2
test. Data are expressed
as mean (), or as median [25–75% interquartile range (IQR)]
for continuous variables, and count (%) for categorical variables.
Conscious sedation for awake craniotomy |813
at University of Basel / A280 UKBB on October 2, 2016http://bja.oxfordjournals.org/Downloaded from

Differences in sedation, pain, and anxiety scores between the
groups were compared using a one-way analysis of variance
(). Repeated-measures  were conducted to assess
variations in MAP, HR, RR, and SpO2over time. For each of the re-
sponses, the interaction between anaesthetic technique and
time was first tested and kept in the model if it reached statistical
significance, or was removed otherwise. An unstructured vari-
ance-covariance structure was used for the within-subject factor.
Least-squares means differences between the groups were com-
pared; associated 95% confidence intervals (CI) and Pvalues are
presented. P<0.05 was considered statistically significant.
Results
Patient characteristics
One-hundred and four patients were screened for study eligibility
between October 2012 and December 2014 (Fig. 1). Fifty-four pa-
tients were excluded before randomization. The remaining 50 pa-
tients were equally randomized to the DEX group (n=25) or the P-R
group (n=25). No participant was lost to follow-up; however, two
patients in the DEX group were excluded from the analysis be-
cause of incorrect allocation in one, and conversion to a general
anaesthetic by surgeon’s request at the start of the procedure in
another.
Baseline patient characteristics and clinical characteristics
are shown in Table 1. There were no differences in patient age,
weight, height, gender, preoperative ASA physical status and
medical co-morbidities, and anaesthesia duration between DEX
and P-R groups. Histological diagnosis of the lesions resected in-
cluded glioma (DEX group, n=12; P-R group, n=11), metastatic
(DEX group, n=6; P-R group, n=10), and other (DEX group, n=5;
P-R group, n=4) (all P>0.05). Arterial lines were inserted for clinical
purposes in four patients (DEX group, n=2; P-R group, n=2). Intrao-
peratively, patients received total doses [mean ()] of fentanyl
[DEX group, 119 (53) µg; P-R group, 89 (39) µg], propofol [DEX
group, 160 (110) mg; P-R group, 596 (531) mg], dexmedetomidine
[DEX group,141 (36) mg], and remifentanil [P-Rgroup, 310 (360) µg].
Outcome variables
Intraoperative brain mapping was successful in all patients [over-
all mean NRS score (): 9.84 (0.48), range 8–10]. There was no dif-
ference between DEX and P-R groups in terms of the ability to
perform brain mapping [mean NRS score (95% CI): DEX group,
10.0 (9.9–10.0) vs P-R group, 9.7 (9.5–10.0), P=0.13].
No difference between groups was found regarding the level
of sedation at the time of mapping [mean OAA/S score (95% CI):
DEX group, 4.1 (3.5–4.7) vs P-R group, 4.3 (3.9–4.7), P=0.51]. The
OAA/S scores were significantly lower in the DEX group at
Assessed for eligibility (n=104)
Lost to follow-up (n=0)
Discontinued intervention (conversion to GA) (n=1)
Lost to follow-up (n=0)
Discontinued intervention (n=0)
Randomized (n=50)
Allocated to P-R group (n=25)
Received allocated intervention (n=25)
Did not receive allocated intervention (n=0)
Analysed (n=25)
Excluded from analysis (n=0)
Analysed (n=23)
Excluded from analysis (n=0)
Allocated to DEX group (n=25)
Received allocated intervention (n=24)
Did not receive allocated intervention (n=1)
Excluded (n=54)
Not meeting inclusion criteria (n=15)
Declined to participate (n=16)
Other reasons (n=23)
Fig 1 CONSORT flow diagram. Fifty subjects were randomized; one subject (DEX group) was eliminated because of incorrect allocation, and one subject (DEX group)
was eliminated because of unexpected intraoperative conversion to a general anaesthetic. DEX group, dexmedetomidine group;GA, general anaesthetic; P-R group,
propofol-remifentanil group.
814 |Goettel et al.
at University of Basel / A280 UKBB on October 2, 2016http://bja.oxfordjournals.org/Downloaded from

intraoperative time points T1–T3 [headpin insertion (P=0.040),
5 min after headpin insertion (P=0.041), and local anaesthetic
infiltration to incision (P=0.018)] (Fig. 2). Arousal times after dis-
continuation of study drug infusion for cortical mapping were
comparable between groups (approximately 5–8min).VASfor
pain was significantly lower in the DEX group at T4 [skin incision
(P=0.026)] and T7 [brain mapping (P=0.031)]. VAS for anxiety was
not different between groups throughout the procedure.
Figure 3shows the time course of haemodynamic an d respira-
tory outcome variables. MAP was significantly lower in the DEX
Table 1 Baseline patient characteristics and clinical characteristics. Data are expressed as mean (SD) or count (%), except for age [mean
(range)] and procedure duration [median (25–75% interquartile range)]. DEX group, dexmedetomidine group; IQR, interquartile range; P-R
group, propofol-remifentanil group
All patients (n=48) P-R group (n=25) DEX group (n=23) Pvalue
Baseline patient characteristics
Age [mean (range); yr] 57.4 (27–88) 53.8 (27–80) 61.4 (36–88) 0.11
Weight [mean (SD); kg] 75.9 (15.2) 73.6 (12.3) 78.4 (17.7) 0.28
Height [mean (SD); cm] 168 (14) 169 (9) 166 (17) 0.98
BMI [mean (); kg m
−2
] 27.4 (6.8) 25.7 (3.9) 29.3 (8.6) 0.07
Gender: male/female [n(%)] 30/18 (62.5/37.5) 16/9 (64/36) 14/9 (60.9/39.1) 0.82
ASA physical status: II/III [n(%)] 8/40 (16.7/83.3) 4/21 (16/84) 4/19 (17.4/82.6) 0.90
Medical co-morbidities [n(%)]
Preoperative seizure 21 (44) 10 (40) 11 (48) 0.59
Respiratory 6 (13) 4 (16) 2 (9) 0.44
Obstructive sleep apnoea 4 (8) 1 (4) 3 (13) 0.26
Cardiac 7 (15) 4 (16) 3 (13) 0.77
Diabetes 3 (6) 1 (4) 2 (9) 0.50
Procedure duration [median (IQR); min] 121 (109–142) 125 (108–177) 115 (108–137) 0.44
0
1
2
3
4
5
6
A
C
B
T0
T1
T2
T3
T4
T5
T6
T7
T8
T9
T10
T11
T12
T0
T1
T2
T3
T4
T5
T6
T7
T8
T9
T10
T11
T12
Modified OAA/S scale
Time course
P-R group DEX group
***
0
1
2
3
4
5
6
7
8
9
10
VAS pain
Time course
P-R group DEX group
*
*
0
1
2
3
4
5
T0
T1
T2
T3
T4
T5
T6
T7
T8
T9
T10
T11
T12
VAS anxiety
Time course
P-R group DEX group
Fig 2 () Modified Observer’s Assessment of Alertness/Sedation (OAA/S) scale (numeric value), () visual analogue scale for pain (VAS pain, numeric value), and ()
visual analogue scale for anxiety (VAS anxiety, numeric value) were assessed at consecutive time points (T0–T12). Study drug infusions were started at T0 and ended
at T10. Results are shown as means (). DEX group, dexmedetom idine group; P-R group, propofol-remifentanil group; T0, intraoperative baseline; T1, headpin
insertion; T2, 5 min after T1; T3, local anaesthetic infiltration to incision; T4, skin incision; T5, craniotomy (bone work); T6, dura opening; T7, brain mapping; T8,
start of tumour resection; T9, 30 min after T8; T10, skin closure; T11, admission to PACU; T12, 120 min after T11. *P<0.05.
Conscious sedation for awake craniotomy |815
at University of Basel / A280 UKBB on October 2, 2016http://bja.oxfordjournals.org/Downloaded from

group at intraoperative time points T6–T8 [dura opening
(P=0.026); brain mapping (P=0.007); start of tumour resection
(P=0.022)] and T11–T12 [admission to PACU (P<0.001); 120 min
after admission to PACU (P=0.004)]. An interaction effect of
treatment group and time was detected for MAP (P=0.044).
Repeated-measures   showed a significantly lower HR
[mean difference (95% CI): −13.8 (−19.3, −8.4) beats min
-1
,
P<0.001] over time in the DEX group. RR was significantly lower
in the P-R group at time points T8 [start of tumour resection
(P=0.030)] and T10 [skin closure (P=0.002)]. There was no differ-
ence SpO
2
between groups throughout the procedure.
Table 2shows the distribution of intraoperative adverse
events. The total incidence of respiratory adverse events with
need for intervention was lower in the DEX group compared
with the P-R group (0 vs 20% respectively, P=0.021). These events
were all short periods of airway obstruction and apnoea, and all
occurred during or immediately after the insertion of head pins,
before draping of the surgical site. Airway obstruction and ap-
noea were quickly treated with jaw thrust and/or brief mask ven-
tilation; the insertion of a naso- or oropharyngeal airway device
was not required at any time. Respiratory adverse events did
not occur in either group during the remaining surgical time.
There was no difference between groups regarding the incidence
of haemodynamic instability, occurrence of a tight brain, new
onset neurological deficits, seizures, excessive pain, psycho-
motor agitation, or nausea and vomiting. Cardiovascular adverse
events, as defined per study protocol, consisted of arterial hypo-
tension treated with ephedrine (n=2) and phenylephrine (n=1),
and arterial hypertension treated with labetalol (n=1) and hydra-
lazine (n=2). One patient (P-R group) developed supraventricular
tachycardia at the end of tumour resection and was treated with
labetalol and esmolol, but required cardioversion in the PACU.
60
70
80
90
100
110
120
T0
T1
T2
T3
T4
T5
T6
T7
T8
T9
T10
T11
T12
T0
T1
T2
T3
T4
T5
T6
T7
T8
T9
T10
T11
T12
T0
T1
T2
T3
T4
T5
T6
T7
T8
T9
T10
T11
T12
T0
T1
T2
T3
T4
T5
T6
T7
T8
T9
T10
T11
T12
Mean arterial pressure (mm Hg)
Time course
AP-R group DEX group
*
*
*
*
**
40
50
60
70
80
90
100
110
Heart rate (beats min–1)
Time course
BP-R group DEX group
*****
**
** ** ** ** **
***
6
8
10
12
14
16
18
20
Respiratory rate (bpm)
Time course
CP-R group DEX group
**
92
93
94
95
96
97
98
99
100
101
102
SpO2 (%)
Time course
DP-R group DEX group
Fig 3 () Mean arterial pressure (MAP, mm Hg), () heart rate (HR, beats min
−1
), () respiratory rate (RR, bpm), and () peripheral oxygen saturation (SpO2, %) were
assessed at consecutive time points (T0–T12). Study drug infusions were started at T0 and ended at T10. Repeated-measures  showed a significantly lower
HR [mean difference (95% CI): −13.8 (−19.3, −8.4) beats min
−1
,P<0.001] over time in the DEX group. Ranges of recorded values for MAP, HR, RR and SpO2in the DEX
group were 50–135 mm Hg, 39–110 beats min
−1
,5–25 bpm and 86–100%, respectively; ranges for MAP, HR, RR and SpO2in the P-R group were 49–136 mm Hg, 46–150
beats min
−1
,6–26 bpm and 90–100%, respectively. Results areshown as means (). DEX group, dexmedetomidine group; P-R group, propofol-remifentanilgroup; T0,
intraoperative baseline;T1, headpin insertion; T2, 5 min after T1; T3, local anaesthetic infiltrationto incision; T4, skin incision; T5, craniotomy (bone work); T6, dura
opening; T7, brain mapping; T8, start of tumour resection; T9, 30 min after T8; T10, skin closure; T11, admission to PACU; T12, 120 min after T11. *P<0.05; **P<0.001;
***P<0.0001.
816 |Goettel et al.
at University of Basel / A280 UKBB on October 2, 2016http://bja.oxfordjournals.org/Downloaded from

One patient (DEX group) experienced a short episode of bradycar-
dia and hypotension (exact values for HR and MAP missing) at the
end of the procedure and was treatedwith atropine. Four patients
in the P-R group developed intraoperative psychomotor agitation
with disinhibition (n=1), or with emotional upset (n=3), of which
one was treated with midazolam. One patient in the DEX group
complained of being ‘too awake’. Seizures occurred in the DEX
group during brain mapping (n=2) and tumour resection (n=1),
and were successfully treated with both cold saline solution ad-
ministered to the brain’s surface and propofol bolus.
Postoperatively, there was no difference in the incidence of
other complications. One patient in the P-R group had a seizure.
The total dose of analgesia administered in the PACU was calcu-
lated by converting the fentanyl, morphine, codeine, and oxy-
codone doses to morphine equivalents.IntheDEXgroup,15
patients (65%) required postoperative analgesia with a mean
() dose of morphine equivalents of 5.6 (3.3) mg; in the P-R
group, 18 patients (72%) with a mean ()doseof7.4(3.8)mg
(P=0.17). Antiemetic medication for prophylactic and/or thera-
peutic purposes was administered in three patients (13%) in the
DEX group and 12 patients (48%) in the P-R group.
The cognitive performance measured at 2 h [mean SPMSQ
score (): DEX group, 0.9 (1.4) vs P-R group, 1.3 (1.8), P=0.43] and
at 24 h [DEX group, 1.5 (1.6) vs P-R group, 1.5 (1.4), P=0.96] was
not different between the two groups, alike the degree of patient
satisfaction and the level of recall of the procedure (Fig. 4). The
final postoperative destination of patients included in the study
did not differ between groups. Thirty-one participants (65%)
were scheduled as outpatients and 14 (29%) as inpatients.
Three patients (6%) that were initially planned for day surgery
were admitted to the hospital after surgery as a result of a new
neurological deficit (DEX group: n=1; P-R group: n=1) and mild
confusion (DEX group: n=1).
Discussion
Dexmedetomidine and propofol-remifentanil-based conscious
sedation, without airway manipulation, during AC for supraten-
torial tumour resection showed similar quality of intraoperative
brain mapping and efficacy of sedation in this prospective, rando-
mized, double-blind, comparative study. The incidence of intra-
and postoperative cardiovascular, neurological, or other adverse
events did not differ between the groups. However, the incidence
of respiratory adverse events was significantly greater in the P-R
group. The levels of perioperative pain and anxiety, patient satis-
faction, and recall were all comparable. Compared with propofol-
remifentanil, dexmedetomidine administration was associated
with a decrease in HR throughout the procedure and a decrease
in MAP during least stimulating surgical time points. However,
the decrease in HR was not greater than 20% from baseline.
The anaesthetic management of an AC using a conscious
sedation technique usually involves a combination of local
anaesthesia to the scalp and i.v. agents to provide sedation,
analgesia, and anxiolysis. Our institutional practice in patients
undergoing AC for tumour surgery is to perform a ‘ring block’in-
filtration of the incision site with bupivacaine, and to provide
concomitant conscious sedation. An alternative to the ‘ring
block’technique is the selective regional anaesthesia to the
nerves that innervate the scalp (‘scalp block’),
28
using different
local anaesthetic agents such as ropivacaine or levobupiva-
caine.
29
Local anaesthetic toxicity is rarely seen in AC.
28
Other an-
aesthetic techniques such as the ‘asleep-awake-asleep’or the
‘asleep-awake’technique, typically involving general anaesthe-
sia and airway management (tracheal intubation or insertion of
a laryngeal mask airway), have been successfully used for AC.
However, when the conscious sedation technique is used, there
is usually no or only minimal manipulation of the airway. Propo-
fol sedation, commonly in combination with a short-acting opi-
oid, is an effective technique for conscious sedation for AC,
914
achieving a high degree of patient satisfaction and acceptance.
10
Other groups recommend the use of target-controlled infusion
(TCI),
30 31
unavailable at our institution, to guide the administra-
tion of i.v. anaesthetics to anticipate the transitions from general
anaesthesia to the awake state during an AC. The use of TCI
modes may also be helpful to prevent respiratory adverse effects
arising from the pharmacological interaction of propofol and
opioid. However, independently of the choice of anaesthetic
technique, AC remains a challenging procedure. The key assets
of an ‘ideal’drug for conscious sedation are a large therapeutic
index and predictable pharmacodynamics to ensure an adequate
level of sedation and analgesia facing the rapid changes of surgi-
cal stimulation, yet permitting the collaboration of the awake pa-
tient in complex intraoperative brain mapping.
Dexmedetomidine produces a cooperative form of sedation,
in which patients easily transition from sleep to wakefulness
and task performance when aroused, and back to sleep when
Table 2 Incidence of intraoperative adverse events. Data are expressed as count (%). CI, confidence interval; DEX group, dexmedetomidine
group; P-R group, propofol-remifentanil group; RR, relative risk
P-R group (n=25) DEX group (n=23) RR 95% CI Pvalue
Respiratory, all events combined [n(%)] 5 (20) 0 10.15 0.59–174.04 0.023
Cardiovascular, all events combined [n(%)] 4 (16) 4 (17) 0.92 0.26–3.26 0.90
Arterial hypertension 2 (8) 1 (4) 1.84 0.18–18.96 0.60
Arterial hypotension 1 (4) 2 (9) 0.46 0.04–4.74 0.50
Cardiac arrhythmia 1 (4) 1 (4) 0.92 0.06–13.87 0.95
Neurological [n(%)]
Tight brain 2 (8) 0 4.62 0.23–91.35 0.17
New neurological deficit 2 (8) 0 4.62 0.23–91.35 0.17
Seizure 0 3 (12) 0.13 0.01–2.42 0.06
Other [n(%)]
Excessive pain 5 (20) 5 (22) 0.92 0.31–2.77 0.88
Psychomotor agitation 4 (16) 1 (4) 3.68 0.44–30.56 0.19
Vomiting 1 (4) 0 2.80 0.12–64.77 0.33
Patients with ≥1 adverse event [n(%)] 13 (52) 12 (52) 0.99 0.58–1.72 0.99
Conscious sedation for awake craniotomy |817
at University of Basel / A280 UKBB on October 2, 2016http://bja.oxfordjournals.org/Downloaded from

not stimulated.
32
Bekker and colleagues
33
first reported the use of
dexmedetomidine in AC in 2001. Subsequent studies evaluating
the influence of dexmedetomidine on the ability to perform in-
traoperative neurologic testing showed inconsistent results.
34–36
One recent case report
37
and several case series of awake crani-
otomies for tumour resection advocate an anaesthetic approach
based on scalp nerve blocks and dexmedetomidine with
38
or
without airway manipulation.
39 40
Another study compared the
combinations of dexmedetomidine and remifentanil to propofol
and remifentanil during AC using an ‘asleep-awake-asleep’tech-
nique involving general anaesthesia with orotracheal intub-
ation.
41
They found both to be effective and safe; however,
there was a shorter arousal time from the sleep state for mapping
with dexmedetomidine. The short arousal times in our study
were likely as a result of relatively low levels of sedation before
brain mapping and the relatively short overall duration of
surgery.
The use of a sole anaesthetic agent may not be sufficient for
all stages of an AC with a conscious sedation technique. The ini-
tial part of the procedure can be very stimulating and painful
with the injection of local anaesthesia, followed by the insertion
of the head pins. During this time the patient may require
additional sedation and analgesia. It is important that the patient
does not experience pain during this part of the procedure.
Therefore, we administered an initial dose of fentanyl to all
patients in our protocol. Also, our past experience had been
that patients were frequently ‘too awake’during periods of dex-
medetomidine sedation alone, hence, we allowed the addition
of rescue medication (propofol bolus), as needed. The opioid-
sparing effects of dexmedetomidine used as an adjunct to anaes-
thesia during the perioperative period are well-documented.
18
But when used as a sole anaesthetic agent, dexmedetomidine
may not offer the desired analgesic effects for all stages of AC,
and thus, may not completely replace opioids.
42 43
A low-dose re-
mifentanil infusion used along with dexmedetomidine may po-
tentially help to achieve successful pain control.
The main safety concerns with conscious sedation in non-in-
tubated patients are airway compromise, hypoventilation and
oxygen desaturation. Most anaesthetic agents used during AC
are associated with some respiratory depression.
12 14
While re-
spiratory adverse events rarely occur when using a technique
that involves intermittent general anaesthesia and invasive air-
way management,
44
spontaneously breathing patients undergo-
ing AC may be at risk for airway obstruction or hypoventilation.
14
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
P-R
group
DEX
group
P-R
group
DEX
group
P-R
group
DEX
group
P-R
group
DEX
group
P-R
group
DEX
group
Very satisfiedABCDE
Mostly satisfied
Indifferent
Quite dissatisfied
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Yes, definitely.
Yes, I think so.
No, I don't think so.
No, definitely not.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Complete
Near complete
Partial
None
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Complete
Near complete
Par tial
None
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Complete
Near complete
Par tial
None
Fig 4 Patient satisfaction and recall of the surgical procedure(awake craniotomy) were assessed at 24 h using Likert scales. In a structuredinterview, patients were
asked to rate their intraoperative experience by answering to the five following questions: () How satisfiedwere you withyour painmanagement and overalllevel of
comfort? () If you were to havesurgery again, would you opt for the same method of management? () Recall of the intraoperativeexperience. () Recall of the level
of intraoperative pain. () Recall of the level of intraoperative discomfort and anxiety. DEX group, dexmedetomidinegroup; P-R group, propofol-remifentanil group.
818 |Goettel et al.
at University of Basel / A280 UKBB on October 2, 2016http://bja.oxfordjournals.org/Downloaded from

In our study, we found an increased incidence of airway and/or
respiratory adverse events within the P-R group. The patient’sre-
spiratory rate increased when propofol was stopped for brain
mapping while it remained constant with dexmedetomidine.
45
In a systematic review of spontaneously breathing subjects re-
ceiving different sedative drugs for sleep endoscopy, all agents
including dexmedetomidine caused some degrees of airway col-
lapse.
46
Thus, dexmedetomidine alone may not cause a decrease
in the respiratory rate or hypoventilation through a central effect
on respiration, but one must be vigilant especially with the add-
ition of other agents, such as opioids and/or propofol, as this may
result in airway obstruction by relaxation of the pharyngeal mus-
cles.
47
For this investigation, we did not measure PaCO
2
and used
EtCO
2
merely for monitoring of RR in spontaneously breathing
patients; however, prolonged alveolar hypoventilation asso-
ciated with clinically important hypercapnia did not seem to
occur in any of our patients.
A decrease in bp and heart rate is the most common cardio-
vascular effect of dexmedetomidine.
48–50
Clinically significant
episodes of hypotension (45%) and bradycardia (14%) have
been associated with dexmedetomidine infusion and may ne-
cessitate medical intervention in 10% and 3% of patients, re-
spectively.
49
The relatively low incidence of haemodynamic
adverse events during conscious sedation for AC found in
both DEX and P-R groups is consistent with findings of previous
studies.
41
Intraoperative seizures have been reported to occur in up to
13% of patients undergoing AC for tumour resection.
51
The risk
is particularly high during brain mapping when electrical cur-
rent is directly applied to the motor cortex (20%).
52
Dexmedeto-
midine has been shown to decrease the seizure threshold in
different animal models.
53 54
However, there are limited
data on its effect on electroencephalographic responses in hu-
mans.
55 56
Several clinical investigations in patients diagnosed
with epilepsy concluded that dexmedetomidine does not reduce
seizure focus activity.
34 57 58
In our study, intraoperative seizures
occurred only in the DEX group (n=3); however, in comparison to
the P-R group, this finding did not reach statistical significance.
Our sample size may have been too small to find any difference.
While the anti-epileptic properties of propofol are known, fur-
ther research should elucidate whether dexmedetomidine has
a direct effect on the seizure threshold (by inhibition of central
noradrenergic transmission), or if the absence of protective
agents such as propofol renders patients more prone to seizures
during AC.
Psychomotor agitation can be an important problem in pa-
tients undergoing complex neurosurgical procedures such as
AC. Disinhibition and lack of cooperation have been described
for low-dose propofol (1.3% of patients)
59
and benzodiazepine
sedation, but do not seem to occur with dexmedetomidine.
32
Ac-
cordingly, we found a trend towards a higher incidence of intrao-
perative psychomotor agitation in the P-R group compared with
the DEX group (P=0.19).
The overall management including the need for analgesia and
incidence of adverse events in the PACU was not different be-
tween the two groups. We were unable to study the need and
the amount of analgesia the patients required after discharge
from PACU as the placement of patients varied. Overall, 58% of
our patients went home on the same day as surgery, which is a
common practice in our institution.
60 61
The SPMSQ was used
as a simple test of memory and cognitive function, and there
were no differences at either time of assessment. Previous stud-
ies have found high satisfaction in patients who underwent an
AC; although recall of intraoperative events varied, most patients
would have the similar technique of anaesthesia if required in
the future.
10 60 62
.
The current study has a number of limitations that should be
considered. Although the patient, surgeon, and study investiga-
tor collecting intraoperative data were blinded to group alloca-
tion, it was not possible to blind the attending anaesthetist
managing the patient for patient safety reasons. The behaviour
of the anaesthetist might have influenced judgement of the sur-
geon and/or the study investigator, and this may be responsible
for bias. The administration of anaesthetic agents being left to
the discretion of the attending anaesthetist may have introduced
additional bias. We acknowledge that our method of comparing
the use of rescue medication in both groups may have been
flawed, as some rescue drug administrations may have stayed
undetected in the P-R group (e.g. when the attending anaesthetist
temporarily increased infusion rates of propofol or remifentanil).
The overall duration of our procedures was relatively short
[median time (IQR): 121 (109–142) min], and the brain mapping
performed was not extensive in terms of examination technique
and duration compared with other studies.
39 41
Thus, the conclu-
sions from our study pertain only to AC for tumour, and may not
be extrapolated to all other neurosurgical procedure performed
as AC, demanding longer procedure times and more complex in-
traoperative neuropsychological testing.
Sample size was calculated only with respect to the primary
outcome measure (NRS of the quality of intraoperative brain
mapping); numerous other outcome variables reported in this
study lack a specific power analysis. Likely, a larger sample size
would be necessary to reveal potential differences between
groups, e.g. in the incidence of adverse events.
We did not utilize a processed EEG-based monitor to evaluate
depth of sedation. Some authors have advocated the use of
bispectral index (BIS) monitoring to guide depth of anaesthesia
during AC and to achieve predictable recovery from general an-
aesthesia, when applying an ‘asleep-awake’protocol.
63
In this
context, an association of TCI modes for drug administration
and BIS may be helpful to reach fast transition times between
anaesthetic states.
30 31
In our study, level of sedation was as-
sessed using the OAA/S scale. Although this is a subjective scor-
ing method based on clinical information, the OAA/S is a reliable
and valid tool with a low inter-rater variability.
26
Previous inves-
tigations have also shown that the OAA/S scale correlates well
with BIS during dexmedetomidine and propofol sedation.
64
In conclusion, the ability to perform intraoperative brain
mappingandtheefficacy of dexmedetomidine was similar
to propofol-remifentanil-based conscious sedation in AC, for su-
pratentorial tumour resection. The use of dexmedetomidine and
propofol-remifentanil during AC was safe. However, dexmedeto-
midine may offer distinct advantages in this indication because
of a lower incidence of respiratory adverse events. Optimal
dose regimen of sedatives and careful vigilance are the keys for
successful conscious sedation during AC.
Ethics committee approval
Ethical approval for this study (ethical committee N° 11-0607-A)
was provided by the University Health Network Research Ethics
Board, 10
th
Floor, Room 1056, 700 University Ave, Toronto, ON,
M5G 1Z5, Canada, Phone: +1 (416) 581-7849, on November 9, 2011.
Approval for this study (control N° 151753) was provided by
Health Canada.
This report describes human research. This study was con-
ducted with written informed consent from the study subjects
and in respect of the revised Declaration of Helsinki.
Conscious sedation for awake craniotomy |819
at University of Basel / A280 UKBB on October 2, 2016http://bja.oxfordjournals.org/Downloaded from

The study was registered on ClinicalTrials.gov (NCT01545297)
before patient enrolment.
This report describes a randomized controlled trial study.
The author states that the report includes the items in the CON-
SORT checklist for randomized controlled trials.
This manuscript was screened for plagiarism with iThenticate.
Authors’contributions
Study design/planning: N.G., P.H.M.
Study conduct: N.G., S.B., L.V., J.M., P.H.M.
Data analysis: N.G.
Writing paper: N.G., P.H.M.
Revising paper: all authors
Supplementary material
Supplementarymaterial is available at British Journal of Anaesthesia
online.
Acknowledgements
The authors wish to thank Ying Qi for her assistance in statistical
analysis and Allison Dwileski for her help in preparation of the
manuscript.
Declaration of interest
None declared.
Funding
Study drug was supplied by Hospira Inc., Lake Forest, USA.
References
1. Szelenyi A, Bello L, Duffau H, et al. Intraoperative electrical
stimulation in awake craniotomy: methodological aspects
of current practice. Neurosurg Focus 2010; 28:E7
2. Taylor MD, Bernstein M. Awake craniotomy with brain map-
ping as the routine surgical approach to treating patients with
supratentorial intraaxial tumors: a prospective trial of 200
cases. J Neurosurg 1999; 90:35–41
3. Serletis D, Bernstein M. Prospective study of awake craniot-
omy used routinely and nonselectively for supratentorial tu-
mors. J Neurosurg 2007; 107:1–6
4. Brown T, Shah AH, Bregy A, et al. Awake craniotomy for
brain tumor resection: the rule rather than the exception?
J Neurosurg Anesthesiol 2013; 25: 240–7
5. Hansen E, Seemann M, Zech N, Doenitz C, Luerding R,
Brawanski A. Awake craniotomies without any sedation:
the awake-awake-awake technique. Acta Neurochir (Wien)
2013; 155: 1417–24
6. Seemann M, Zech N, Graf B, Hansen E. Anesthesiological
management of awake craniotomy : Asleep-awake-asleep
technique or without sedation. Anaesthesist 2015; 64: 128–36
7. Blanshard HJ, Chung F, Manninen PH, Taylor MD,
Bernstein M. Awake craniotomy for removal of intracranial
tumor: considerations for early discharge. Anesth Analg
2001; 92:89–94
8. Conte V, Magni L, Songa V, et al. Analysis of propofol/remifen-
tanil infusion protocol for tumor surgery with intraoperative
brain mapping. J Neurosurg Anesthesiol 2010; 22: 119–27
9. Berkenstadt H, Perel A, Hadani M, Unofrievich I, Ram Z. Mon-
itored anesthesia care using remifentanil and propofol for
awake craniotomy. J Neurosurg Anesthesiol 2001; 13: 246–9
10. Manninen PH, Balki M, Lukitto K, Bernstein M. Patient satis-
faction with awake craniotomy for tumor surgery: a compari-
son of remifentanil and fentanyl in conjunction with
propofol. Anesth Analg 2006; 102: 237–42
11. Johnson KB, Egan TD. Remifentanil and propofol combin-
ation for awake craniotomy: case report with pharmacoki-
netic simulations. J Neurosurg Anesthesiol 1998; 10:25–9
12. Skucas AP, Artru AA. Anesthetic complications of awake cra-
niotomies for epilepsy surgery. Anesth Analg 2006; 102:882–7
13. Herrick IA, Craen RA, Gelb AW, et al. Propofol sedation during
awake craniotomy for seizures: patient-controlled adminis-
tration versus neurolept analgesia. Anesth Analg 1997; 84:
1285–91
14. Sarang A, Dinsmore J. Anaesthesia for awake craniotomy–
evolution of a technique that facilitates awake neurological
testing. Br J Anaesth 2003; 90: 161–5
15. Bhana N, Goa KL, McClellan KJ. Dexmedetomidine. Drugs
2000; 59: 263–8
16. Kamibayashi T, Maze M. Clinical uses of alpha2 -adrenergic
agonists. Anesthesiology 2000; 93: 1345–9
17. Karol MD, Maze M. Pharmacokinetics and interaction
pharmacodynamics of dexmedetomidine in humans. Best
Practice & Research Clinical Anaesthesiology 2000; 14: 261–9
18. Arain SR, Ruehlow RM, Uhrich TD, Ebert TJ. The efficacy of
dexmedetomidine versus morphine for postoperative anal-
gesia after major inpatient surgery. Anesth Analg 2004; 98:
153–8
19. Hsu YW, Cortinez LI, Robertson KM, et al. Dexmedetomidine
pharmacodynamics: part I: crossover comparison of the re-
spiratory effects of dexmedetomidine and remifentanil in
healthy volunteers. Anesthesiology 2004; 101: 1066–76
20.DyckJB,MazeM,HaackC,AzarnoffDL,VuorilehtoL,
Shafer SL. Computer-controlled infusion of intravenous dex-
medetomidine hydrochloride in adult human volunteers.
Anesthesiology 1993; 78: 821–8
21. Venn RM, Karol MD, Grounds RM. Pharmacokinetics of dex-
medetomidine infusions for sedation of postoperative pa-
tients requiring intensive caret. Br J Anaesth 2002; 88: 669–75
22. Talke P, Richardson CA, Scheinin M, Fisher DM. Postoperative
pharmacokinetics and sympatholytic effects of dexmedeto-
midine. Anesth Analg 1997; 85: 1136–42
23. Iirola T, Ihmsen H, Laitio R, et al. Population pharmacokinet-
ics of dexmedetomidine during long-term sedation in inten-
sive care patients. Br J Anaesth 2012; 108: 460–8
24. Nelson LE, Lu J, Guo T, Saper CB, Franks NP, Maze M. The
alpha2-adrenoceptor agonist dexmedetomidine converges
on an endogenous sleep-promoting pathway to exert its
sedative effects. Anesthesiology 2003; 98: 428–36
25. Maze M, Scarfini C, Cavaliere F. New agents for sedation in the
intensive care unit. Crit Care Clin 2001; 17: 881–97
26. Chernik DA, Gillings D, Laine H, et al. Validity and reliability of
the Observer’s Assessment of Alertness/Sedation Scale:
study with intravenous midazolam. J Clin Psychopharmacol
1990; 10: 244–51
27. Pfeiffer E. A short portable mental status questionnaire for
the assessment of organic brain deficit in elderly patients.
J Am Geriatr Soc 1975; 23: 433–41
28. Osborn I, Sebeo J. ‘Scalp block’during craniotomy: a classic
technique revisited. J Neurosurg Anesthesiol 2010; 22: 187–94
29. Costello TG, Cormack JR. Anaesthesia for awake craniotomy:
a modern approach. J Clin Neurosci 2004; 11:16–9
820 |Goettel et al.
at University of Basel / A280 UKBB on October 2, 2016http://bja.oxfordjournals.org/Downloaded from

30. Hans P, Bonhomme V, Born JD, Maertens de Noordhoudt A,
Brichant JF, Dewandre PY. Target-controlled infusion of pro-
pofol and remifentanil combined with bispectral index mon-
itoring for awake craniotomy. Anaesthesia 2000; 55: 255–9
31. Lobo F, Beiras A. Propofol and remifentanil effect-site con-
centrations estimated by pharmacokinetic simulation and
bispectral index monitoring during craniotomy with intrao-
perative awakening for brain tumor resection. J Neurosurg
Anesthesiol 2007; 19: 183–9
32. Bekker A, Sturaitis MK. Dexmedetomidine for neurological
surgery. Neurosurgery 2005; 57:1–10
33. Bekker AY, Kaufman B, Samir H, Doyle W. The use of dexme-
detomidine infusion for awake craniotomy. Anesth Analg
2001; 92: 1251–3
34. Souter MJ, Rozet I, Ojemann JG, et al. Dexmedetomidine sed-
ation during awake craniotomy for seizure resection: effects
on electrocorticography. J Neurosurg Anesthesiol 2007; 19:
38–44
35. Bustillo MA, Lazar RM, Finck AD, et al. Dexmedetomidine may
impair cognitive testing during endovascular embolization of
cerebral arteriovenous malformations: a retrospective case
report series. J Neurosurg Anesthesiol 2002; 14: 209–12
36. Mack PF, Perrine K, Kobylarz E, Schwartz TH, Lien CA. Dexme-
detomidine and neurocognitive testing in awake craniotomy.
J Neurosurg Anesthesiol 2004; 16:20–5
37. Kallapur BG, Bhosale R. Use of dexmedetomidine infusion in
anaesthesia for awake craniotomy. Indian J Anaesth 2012; 56:
413–5
38. Chung YH, Park S, Kim WH, Chung IS, Lee JJ. Anesthetic man-
agement of awake craniotomy with laryngeal mask airway
and dexmedetomidine in risky patients. Korean J Anesthesiol
2012; 63: 573–5
39. Garavaglia MM, Das S, Cusimano MD, et al. Anesthetic
approach to high-risk patients and prolonged awake craniot-
omy using dexmedetomidine and scalp block. J Neurosurg
Anesthesiol 2014; 26: 226–33
40. Mohd Nazaruddin WH, Mohd Fahmi L, Laila AM, Zamzuri I,
Abdul Rahman IZ, Hardy MZ. Awake Craniotomy: A Case
Series of Anaesthetic Management using a Combination of
Scalp Block, Dexmedetomidine and Remifentanil in Hospital
Universiti Sains Malaysia. Med J Malaysia 2013; 68:64–6
41. Shen SL, Zheng JY, Zhang J, et al. Comparison of dexmedeto-
midine and propofol for conscious sedation in awake craniot-
omy: a prospective, double-blind, randomized, and
controlled clinical trial. Ann Pharmacother 2013; 47: 1391–9
42. Jaakola ML, Salonen M, Lehtinen R, Scheinin H. The analgesic
action of dexmedetomidine–a novel alpha 2-adrenoceptor
agonist–in healthy volunteers. Pain 1991; 46: 281–5
43. Paris A, Tonner PH. Dexmedetomidine in anaesthesia. Curr
Opin Anaesthesiol 2005; 18: 412–8
44. Deras P, Moulinie G, Maldonado IL, Moritz-Gasser S, Duffau H,
Bertram L. Intermittent general anesthesia with controlled
ventilation for asleep-awake-asleep brain surgery: a prospect-
ive series of 140 gliomas in eloquent areas. Neurosurgery 2012;
71:764–71
45. Olofsen E, Boom M, Nieuwenhuijs D, et al. Modeling the non-
steady state respiratory effects of remifentanil in awake and
propofol-sedated healthy volunteers. Anesthesiology 2010;
112: 1382–95
46. Ehsan Z, Mahmoud M, Shott SR, Amin RS, Ishman SL. The
effects of Anesthesia and opioids on the upper airway: A sys-
tematic review. Laryngoscope 2016; 126: 270–84
47. Ramsay MA, Luterman DL. Dexmedetomidine as a total intra-
venous anesthetic agent. Anesthesiology 2004; 101: 787–90
48. Ebert TJ, Hall JE, Barney JA, Uhrich TD, Colinco MD. The effects
of increasing plasma concentrations of dexmedetomidine in
humans. Anesthesiology 2000; 93: 382–94
49. Candiotti KA, Bergese SD, Bokesch PM, et al. Monitored anes-
thesia care with dexmedetomidine: a prospective, rando-
mized, double-blind, multicenter trial. Anesth Analg 2010;
110:47–56
50. Bergese SD, Candiotti KA, Bokesch PM, et al. A Phase IIIb, ran-
domized, double-blind, placebo-controlled, multicenter
study evaluating the safety and efficacy of dexmedetomidine
for sedation during awake fiberoptic intubation. Am J Ther
2010; 17: 586–95
51. Nossek E, Matot I, Shahar T, et al. Intraoperative seizures dur-
ing awake craniotomy: incidence and consequences: analysis
of 477 patients. Neurosurgery 2013; 73: 135–40
52. Bonhomme V, Franssen C, Hans P. Awake craniotomy. Eur J
Anaesthesiol 2009; 26: 906–12
53. Mirski MA, Rossell LA, McPherson RW, Traystman RJ. Dexme-
detomidine decreases seizure threshold in a rat model of
experimental generalized epilepsy. Anesthesiology 1994; 81:
1422–8
54. Miyazaki Y, Adachi T, Kurata J, Utsumi J, Shichino T,
Segawa H. Dexmedetomidine reduces seizure threshold
during enflurane anaesthesia in cats. Br J Anaesth 1999; 82:
935–7
55.HuupponenE,MaksimowA,LapinlampiP,et al. Electro-
encephalogram spindle activity during dexmedetomidine
sedation and physiological sleep. Acta Anaesthesiol Scand
2008; 52: 289–94
56. Akeju O, Pavone KJ, Westover MB, et al. A comparison of pro-
pofol- and dexmedetomidine-induced electroencephalo-
gram dynamics using spectral and coherence analysis.
Anesthesiology 2014; 121: 978–89
57. Talke P, Stapelfeldt C, Garcia P. Dexmedetomidine does not
reduce epileptiform discharges in adults with epilepsy.
J Neurosurg Anesthesiol 2007; 19: 195–9
58. Oda Y, Toriyama S, Tanaka K, et al. The effect of dexmedeto-
midine on electrocorticography in patients with temporal
lobe epilepsy under sevoflurane anesthesia. Anesth Analg
2007; 105: 1272–7
59. McLeskey CH, Walawander CA, Nahrwold ML, et al. Adverse
events in a multicenter phase IV study of propofol: evaluation
by anesthesiologists and postanesthesia care unit nurses.
Anesth Analg 1993; 77:S3–9
60. Khu KJ, Doglietto F, Radovanovic I, et al. Patients perceptions
of awake and outpatient craniotomy for brain tumor: a quali-
tative study. J Neurosurg 2010; 112: 1056–60
61. Purzner T, Purzner J, Massicotte EM, Bernstein M. Outpatient
brain tumor surgery and spinal decompression: a prospective
study of 1003 patients. Neurosurgery 2011; 69: 119–26
62. Milian M, Tatagiba M, Feigl GC. Patient response to awake cra-
niotomy - a summary overview. Acta Neurochir (Wien) 2014;
156: 1063–70
63. Conte V, L’Acqua C, Rotelli S, Stocchetti N. Bispectral index
during asleep-awake craniotomies. J Neurosurg Anesthesiol
2013; 25: 279–84
64. Kasuya Y, Govinda R, Rauch S, Mascha EJ, Sessler DI, Turan A.
The correlation between bispectral index and observational
sedation scale in volunteers sedated with dexmedetomidine
and propofol. Anesth Analg 2009; 109: 1811–5
Handling editor: A. R. Absalom
Conscious sedation for awake craniotomy |821
at University of Basel / A280 UKBB on October 2, 2016http://bja.oxfordjournals.org/Downloaded from