Você está na página 1de 10


Dexmedetomidine for prevention of delirium in elderly

patients after non-cardiac surgery: a randomised, double-blind,
placebo-controlled trial
Xian Su, Zhao-Ting Meng, Xin-Hai Wu, Fan Cui, Hong-Liang Li, Dong-Xin Wang, Xi Zhu, Sai-Nan Zhu, Mervyn Maze, Daqing Ma

Background Delirium is a postoperative complication that occurs frequently in patients older than 65 years, and Published Online
presages adverse outcomes. We investigated whether prophylactic low-dose dexmedetomidine, a highly selective August 16, 2015
2 adrenoceptor agonist, could safely decrease the incidence of delirium in elderly patients after non-cardiac surgery. S0140-6736(16)30580-3
See Online/Comment
Methods We did this randomised, double-blind, placebo-controlled trial in two tertiary-care hospitals in Beijing, http://dx.doi.org/10.1016/
China. We enrolled patients aged 65 years or older, who were admitted to intensive care units after non-cardiac S0140-6736(16)31353-8
surgery, with informed consent. We used a computer-generated randomisation sequence (in a 1:1 ratio) to randomly Department of
assign patients to receive either intravenous dexmedetomidine (01 g/kg per h, from intensive care unit admission Anaesthesiology and Critical
Care Medicine (X Su MD,
on the day of surgery until 0800 h on postoperative day 1), or placebo (intravenous normal saline). Participants, care
Z-T Meng MD, X-H Wu MD,
providers, and investigators were all masked to group assignment. The primary endpoint was the incidence of F Cui MD, Prof D-X Wang MD),
delirium, assessed twice daily with the Confusion Assessment Method for intensive care units during the rst Department of Biostatistics
7 postoperative days. Analyses were done by intention-to-treat and safety populations. This study is registered with (S N Zhu MS), Peking University
First Hospital, Beijing, China;
Chinese Clinical Trial Registry, www.chictr.org.cn, number ChiCTR-TRC-10000802.
Department of Critical Care
Medicine, Peking University
Findings Between Aug 17, 2011, and Nov 20, 2013, of 2016 patients assessed, 700 were randomly assigned to receive Third Hospital, Beijing, China
either placebo (n=350) or dexmedetomidine (n=350). The incidence of postoperative delirium was signicantly lower (H-L Li MD, Prof X Zhu MD);
Department of Anesthesia and
in the dexmedetomidine group (32 [9%] of 350 patients) than in the placebo group (79 [23%] of 350 patients; odds Perioperative Care, University
ratio [OR] 035, 95% CI 022054; p<00001). Regarding safety, the incidence of hypertension was higher with of California, San Francisco, CA,
placebo (62 [18%] of 350 patients) than with dexmedetomidine (34 [10%] of 350 patients; 050, 032078; p=0002). USA (Prof M Maze MBChB); and
Tachycardia was also higher in patients given placebo (48 [14%] of 350 patients) than in patients given dexmedetomidine Section of Anaesthetics,
Pain Management and
(23 [7%] of 350 patients; 044, 026075; p=0002). Occurrence of hypotension and bradycardia did not dier Intensive Care, Department of
between groups. Surgery and Cancer, Imperial
College London, Chelsea and
Interpretation For patients aged over 65 years who are admitted to the intensive care unit after non-cardiac surgery, Westminster Hospital, London,
UK (Prof D Ma MD)
prophylactic low-dose dexmedetomidine signicantly decreases the occurrence of delirium during the rst 7 days
Correspondence to:
after surgery. The therapy is safe. Prof Dong-Xin Wang,
Department of Anaesthesiology
Funding Braun Anaesthesia Scientic Research Fund and Wu Jieping Medical Foundation, Beijing, China. Study and Critical Care Medicine,
drugs were manufactured and supplied by Jiangsu Hengrui Medicine Co, Ltd, Jiangsu, China. Peking University First Hospital,
Beijing 100034, China
Introduction mechanically ventilated patients in the intensive care unit or
A systematic review1 revealed that postoperative delirium (ICU),6 where its use is associated with a decreased Prof Daqing Ma, Anaesthetics,
occurs in 1151% of patients after surgery, and its prevalence of delirium when compared with other Pain Medicine and Intensive
prevalence increases with age. The occurrence of sedatives.7,8 However, in each of these delirium-sparing Care, Department of Surgery and
Cancer, Faculty of Medicine,
delirium is associated with increased morbidity and studies,79 dexmedetomidine was compared with an active
Imperial College London, Chelsea
mortality, prolonged hospital stay, worse functional sedative drug that modulates the -aminobutyric-acid and Westminster Hospital,
recovery, and long-term decline in cognitive function.1,2 type A (GABAA) receptors. These modulators of GABAA London, UK
In patients admitted to hospital, around 3040% of receptors, exemplied by benzodiazepines, could increase d.ma@imperial.ac.uk

delirium cases are thought to be attributable to modiable the prevalence of delirium.10 Another plausible
risk factors, and are therefore preventable.3 Various explanation is that dexmedetomidine does not prevent
approaches aimed at minimising the inuence of risk the occurrence of delirium, but also does not increase the
factors in medical patients have not improved outcomes, prevalence of delirium as do modulators of the GABAA
and there are no conclusive studies that support receptors. Furthermore, the targeted patients were
pharmacological prophylaxis.4 mechanically ventilated, which itself increases the risk
Dexmedetomidine is a highly selective 2 adrenoreceptor of delirium.11 Therefore, it is not clear whether
agonist that provides anxiolysis, sedation, and modest dexmedetomidine has preventive eects against delirium
analgesia with minimal respiratory depression.5 in other patient populations, including non-ventilated
Dexmedetomidine is increasingly used for sedation in patients. Lastly, the sedative dose of dexmedetomidine

www.thelancet.com Published online August 16, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30580-3 1


Research in context
Evidence before this study The meta-analysis publication suggested that
We searched PubMed between Jan 1, 2001, and Dec 31, 2015, dexmedetomidine use in perioperative conditions or for
with the terms of dexmedetomidine, postoperative intesive care unit (ICU) sedation is associated with a low risk of
delirium, and elderly, and then limited to either randomized neurocognitive dysfunction.
controlled trial or meta-analysis. We identied seven small Added value of this study
sample size randomised trials (three published in English and To our knowledge, this study is the rst to suggest that
four in Chinese) and one meta-analysis. Dexmedetomidine was prophylactic infusion of low-dose dexmedetomidine, an
administered in ve trials in a relatively high dose for sedation 2 adrenoceptor agonist sedative, signicantly decreases the
in postoperative patients. Of those, the primary endpoint was prevalence of postoperative delirium without increase in
not the incidence or prevalence of delirium in three trials; in two adverse events.
trials published in Chinese, dexmedetomidine decreased the
incidence of delirium during the rst 3 days after oral cancer Implications of all available evidence
surgery in one trial while, in the other, patient-controlled The sub-sedative dose of dexmedetomidine can be safely used
analgesia supplemented with dexmedetomidine, given to for elderly ICU patients after surgery, both with and without
elderly patients after spinal surgery, did not reduce the endotracheal intubation, to reduce the likelihood of
incidence of delirium during the rst 3 postoperative days. postoperative delirium.

used in the earlier studies9 was associated with an patients whose competence was established by their
increase in hypotension or bradycardia, which limits a accurate orientation for time, place, and person, and
wider clinical application.6 Because dexmedetomidine understanding of the recruiters description of the trial, or
induces haemodynamic changes in a dose-dependent otherwise from their next of kin or their legal representative
pattern,12 it is important to dene whether a lower dose (appendix).
than that used in other studies is still benecial in
decreasing delirium with fewer haemodynamic changes. Patients
Sleep disturbances are common in postoperative We screened potential participants on admission to the
patients, especially in those who are admitted to the ICU ICU. The inclusion criteria were patients aged 65 years or
after major surgery,13 and poor sleep is associated with a older who underwent elective non-cardiac surgery under
higher prevalence of postoperative delirium.14 Results of a general anaesthesia and were admitted to the ICU after
2014 study15 showed that night-time infusion of sedative surgery before 2000 h. Patients were excluded if they met
dose dexmedetomidine improved sleep quality in any of the following criteria: preoperative history of
mechanically ventilated ICU patients. With these results schizophrenia, epilepsy, Parkinsonism, or myasthenia
in mind, we did a feasibility study to test our hypothesis gravis; inability to communicate in the preoperative period
that low-dose dexmedetomidine infusion at a rate of (coma, profound dementia, or language barrier); brain
01 g/kg per h could be benecial for patients sleep and injury or neurosurgery; known preoperative left ventricular
beyond. We found that prophylactic infusion of low-dose ejection fraction less than 30%, sick sinus syndrome, severe
dexmedetomidine improved the overall sleep quality sinus bradycardia (<50 beats per min [bpm]), or second-
measured by polysomnography and subjective degree or greater atrioventricular block without pacemaker;
assessment. This nding encouraged us to do a serious hepatic dysfunction (Child-Pugh class C); serious
randomised controlled trial with a large sample size to renal dysfunction (undergoing dialysis before surgery); or
investigate whether prophylactic intravenous infusion of low likelihood of survival for more than 24 h.
low-dose dexmedetomidine decreases delirium in patients
aged over 65 years (hereafter referred to as elderly patients) Randomisation and masking
admitted to the ICU after non-cardiac surgery. A biostatistician, who was independent of data
management and statistical analyses, generated random
Methods numbers (in a 1:1 ratio) using the SAS 9.2 software (SAS
Study design Institute, Cary, NC). The results of randomisation were
We did a randomised, double-blind, parallel-arm placebo- sealed in sequentially numbered envelopes and stored at
controlled trial in the ICUs of Peking University First the site of investigation until the end of the study.
Hospital and Peking University Third Hospital in Beijing, During the study period, consecutively recruited ICU
China. The study was designed to assess the superiority of patients were randomly assigned to receive either
See Online for appendix the intervention. The study protocol (appendix) was dexmedetomidine or placebo (normal saline). A study
approved by the local Clinical Research Ethics Committees nurse administered the study drugs according to the
(2011[10]). We obtained written informed consent from randomisation sequence. Study personnel, health-care

2 www.thelancet.com Published online August 16, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30580-3


team members, and patients were masked to the severe agitation (RASS score of +3 or more) that was
treatment group assignment throughout the study unresponsive to non-pharmacological therapy.19 Enrolled
period. In an emergency (eg, unexpected, rapid patients were not to be given open-label dexmedetomidine;
deterioration in the patients clinical status), intensivists scopolamine and penehyclidine were prohibited; atropine
could request unmasking of the treatment allocation, or was used only for the purpose of reversing bradycardia.
adjust or interrupt study drug infusion if necessary. No ICU discharge was decided by the responsible intensivists;
unmasking occurred. These situations were documented, hospital discharge was decided by the responsible
although analyses were done on the intention-to-treat surgeons. Time of actual discharge was recorded.
Procedures Outcome assessment was done by research members
Study drugs (dexmedetomidine hydrochloride 200 g/2 mL who were trained before the study and not involved in the
and normal saline 2 mL) were provided as clear aqueous clinical care of patients. The primary endpoint was the
solutions in the same 3 mL bottles (manufactured by incidence of delirium in the rst 7 days after surgery. The
Jiangsu Hengrui Medicine Co, Ltd, Jiangsu, China) and rst assessment of postoperative (also referred to as
dispensed according to the randomisation results by a interval) delirium was done around 24 h after surgery;20 we
pharmacist who did not participate in the rest of the study. selected the timing of the rst assessment to avoid
The study drugs were diluted with normal saline to 50 mL diagnosing emergence delirium that can occur
(ie, dexmedetomidine hydrochloride nal concentration immediately after general anaesthesia and is not
was 4 g/mL) before administration. associated with adverse outcomes.21,22 Twice daily (in the
For patients who were not intubated, study drugs were morning from 0800 h to 1000 h and in the evening from
given as a continuous intravenous infusion at a rate of 1800 h to 2000 h) until the seventh day after surgery, we
0025 mL/kg per h (01 g/kg per h of dexmedetomidine assessed delirium by the Confusion Assessment Method
in the treatment group) from study recruitment on the for the ICU (CAM-ICU); (appendix),23 which has been
day of surgery (usually within 1 h after ICU admission) validated in Chinese patients in the ICU setting24 and the
until 0800 h on the rst day after surgery. For those who feasibility of which had been established in our other
were intubated and mechanically ventilated, the study studies.25,26 CAM-ICU addresses the four features of
drug infusion was started only after sedative (propofol or delirium, namely, acute onset of mental status changes or
midazolam) administration was titrated to a Richmond a uctuating course, inattention, disorganised thinking,
Agitation Sedation Scale (RASS)16 of 2 or higher and altered level of consciousness. To achieve the
(assessed hourly). diagnosis of delirium, a patient had to display acute onset
Postoperative analgesia was given with patient-controlled of mental status changes or uctuating course and
intravenous or epidural analgesia. For patients who inattention, with either disorganised thinking or altered
did not receive patient-controlled analgesia or those who level of consciousness. Immediately before assessing
required analgesia in addition to that provided delirium, sedation or agitation was assessed using RASS.
from patient-controlled dispensers, morphine or non- If the patient was too deeply sedated or unarousable
steroid anti-inammatory drugs (urbiprofen axetil) were (RASS 4 or 5), delirium assessment was aborted and the
given via intravenous infusion or bolus injection. patient was recorded as comatose. If RASS was greater
Mechanically ventilated patients were sedated with than 4 (3 to +4), delirium was assessed by use of the
propofol or midazolam via intravenous infusion or bolus CAM-ICU. Patients with delirium were classied into
injection, and morphine as necessary, titrated to achieve three motoric subtypes. Hyperactive delirium was dened
an RASS between 2 and +1 (assessed every 4 h). Daily when RASS was consistently positive (+1 to +4); hypoactive
sedation interruption was done for those who were not delirium was dened when RASS was consistently neutral
extubated by the morning of postoperative day 1. Patients or negative (3 to 0); and mixed delirium was dened
were extubated when they met the following three criteria: when some RASS scores were positive (+1 to +4) and some
adequate gas exchange during a spontaneous breathing RASS scores were neutral or negative (3 to 0).27 For
trial, stable haemodynamic status (20% over or under patients who were discharged or died within 7 days after
baseline), and a level of consciousness associated with surgery, the results of the last delirium assessment were
reexes that protect the airway. considered the results of the missing data. These patients
Several approaches to reduce the occurrence of delirium were excluded when calculating daily prevalence of
were instituted as part of standard operating procedures delirium in a post-hoc analysis.
for patients in the ICU, including repeated reorientation, Secondary endpoints included time to extubation (from
cognitive stimulation, early mobilisation, sleep-promotion ICU admission to extubation), length of stay in the ICU
strategies, hearing or vision aids, and timely correction of (from ICU admission to ICU discharge), length of stay in
dehydration.17 Patients who developed postoperative the hospital after surgery (from day of surgery to hospital
delirium were rst given non-pharmacological strategies.18 discharge), occurrence of non-delirium postoperative
Haloperidol treatment was administered to those with complications, and all-cause 30-day mortality. Non-delirium

www.thelancet.com Published online August 16, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30580-3 3


a decrease of more than 5% (absolute value) from

2016 patients assessed for eligibility baseline. Intervention for bradycardia, tachycardia, and
1181 patients excluded
636 aged less than 65 years hypertension included adjustment of study drug infusion
94 non-surgical patients or administration of medication, or both. Intervention for
183 non-general anaesthesia
147 neurosurgery hypotension included adjustment of study drug infusion,
70 required dialysis preoperatively intravenous uid bolus, or administration of medication.
28 preoperative coma
17 preoperative LVEF<30%
Intervention for hypoxaemia included administration of
6 preoperative Parkinsonism oxygen (for patients without endotracheal intubation),
835 patients eligible
adjustment of ventilator setting (for patients with
endotracheal intubation), or physiotherapy.
135 refused to participate
Patients were followed up weekly after the rst week
until 30 days after surgery. All-cause 30-day mortality was
700 randomised
recorded (appendix).

Statistical analysis
350 assigned to receive dexmedetomidine 350 assigned to receive placebo In our study, the incidence of postoperative delirium in a
comparable patient population was 28%. In previous
studies,8,9 the incidence of delirium was reduced by
83 discharged within 7 days 60 discharged within 7 days
32 study drug infusion modified 16 study drug infusion modified roughly a third when dexmedetomidine was used in the
1 died within 7 days 1 died within 7 days ICU for sedating mechanically ventilated patients. Thus,
we assumed that the incidence of delirium would be
350 included in final ITT analyses 350 included in final ITT analyses
reduced by a third in the dexmedetomidine group in this
study. With signicance set at 005 and power set at 80%,
the sample size required to detect dierences was
Figure 1: Trial prole
ITT analyses included all randomised patients in the groups to which they were randomly assigned. ITT=intention-to- 656 patients, calculated with the Stata 10.0 software
treat. LVEF=left ventricular injection fraction. (StataCorp LP, College Station, TX, USA). Taking into
account a loss-to-follow-up rate of about 6%, we planned
complications were generally dened as medical events to enrol 700 patients.
other than delirium that required therapeutic intervention Numeric variables were analysed by use of an unpaired
and occurred within 30 days after surgery (appendix). t test or Mann-Whitney u test. Categorical variables were
Additional prespecied endpoints included postoperative analysed with the test, continuity correction test or
pain intensity and subjective sleep quality. Pain intensity likelihood ratio test. The dierence (and 95% CI for
both at rest and with movement was assessed by use of the the dierence) between two medians was calculated with
Numeric Rating Scale (NRS, an 11 point scale where 0 the Hodges-Lehmann estimator. Time to event results
indicated no pain and 10 indicated the worst possible pain) were calculated with the Kaplan-Meier estimator, with
at 3 h, 6 h, and 24 h after surgery. Subjective sleep quality dierences between groups assessed by the log-rank test.
was assessed by use of the NRS as well (an 11 point scale Number needed to treat was estimated for the primary
where 0 indicated the best possible sleep and 10 indicated endpoint during a 7-day follow-up period. Because
the worst possible sleep)28 at 0800 h on the rst, second, management of patients with or without endotracheal
and third days after surgery. Assessments of pain and intubation on ICU admission is dierent, post-hoc
sleep were only done if the RASS score was more than 4 subgroup analyses were also done.
(3 to +4). We analysed outcome data and safety in the
Adverse events were monitored until 24 h after surgery intention-to-treat population. We also did per-protocol
or until resolution of the event. Bradycardia was dened analysis for the primary endpoint. We did not do an interim
as heart rate less than 55 bpm or a decrease of more than analysis. Statistical analyses were done on SPSS 14.0 software
20% from baseline (in case of a baseline value [before (SPSS, Chicago, IL) and SAS 9.2 software (SAS Institute,
study drug infusion] less than 69 bpm). Hypotension was Cary, NC) with two-tailed tests wherever appropriate and
dened as systolic blood pressure less than 95 mm Hg or p values less than 005 were considered to be of statistical
a decrease of more than 20% from baseline (in case of a signicance. The Clinical Research Ethics Committee from
baseline value less than 119 mm Hg). Tachycardia was Peking University First Hospital was involved in overseeing
dened as heart rate greater than 100 bpm or an increase the data. The study is registered with www.chictr.org.cn,
of more than 20% from baseline (in case of a baseline number ChiCTR-TRC-10000802.
value greater than 83 bpm). Hypertension was dened as
systolic blood pressure greater than 160 mm Hg or an Role of the funding source
increase of more than 20% from baseline (in case of a The study sponsors had no role in study design, in the
baseline value greater than 133 mm Hg). Hypoxaemia collection, analysis, and interpretation of data, or in the
was dened as pulse oxygen saturation less than 90% or writing of the report. The corresponding authors had full

4 www.thelancet.com Published online August 16, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30580-3


access to all of the data in the study and take responsibility

Placebo group (n=350) Dexmedetomidine
for the integrity of the data and the accuracy of the data group (n=350)
analysis. The corresponding authors have nal
Benzodiazepine use in preoperative night 39 (111%) 40 (114%)
responsibility for the decision to submit for publication.
Type of anaesthesia
General 290 (829%) 288 (823%)
Combined epidural-general 60 (171%) 62 (177%)
Between Aug 17, 2011, and Nov 20, 2013, 2016 patients
Intraoperative medication
were screened for study participation; of these,
Midazolam 173 (494%) 153 (437%)
700 patients were enrolled into the study and randomly
Dexamethasone* 260 (743%) 262 (749%)
assigned to receive either dexmedetomidine (n=350) or
placebo (n=350); (gure 1). During the study period, Other glucocorticoids 80 (229%) 77 (220%)

there were no lapses in the blinding. Study drug infusion Atropine 36 (103%) 38 (109%)

was modied in 48 patients because of adverse events. Duration of anaesthesia (min) 320 (156) 307 (135)
Delirium assessment was completed in all patients on Surgery for malignant tumour 252 (720%) 274 (783%)
the rst day of ICU admission. No assessment was Type of surgery
aborted because of deep sedation. 143 patients were Intra-abdominal 235 (671%) 240 (686%)
discharged from the hospital within 7 days after surgery. Intra-thoracic 56 (160%) 64 (183%)
Two patients died, both on postoperative day 2 (one Spinal and extremital 24 (69%) 12 (34%)
patient in each group). All patients were included in the Supercial and transurethral 35 (100%) 34 (97%)
nal intention-to-treat analyses (gure 1). The nal visit Duration of surgery (min) 238 (148) 219 (124)
of the last randomised patient was done on Dec 20, 2013. Estimated blood loss during surgery (mL) 250 (100600) 200 (100500)
Overall, the two groups were well matched for baseline Blood transfusion during surgery 67 (191%) 47 (134%)
and perioperative variables, except that the percentage of Total intraoperative infusion (mL) 2600 (16004100) 2400 (16003600)
patients with preoperative renal dysfunction (serum Patients with endotracheal intubation on ICU admission 191 (546%) 191 (546%)
creatinine greater than 177 mol/L) was lower in the APACHE II score on ICU admission (score) 106 (39) 102 (33)
dexmedetomidine group than in the placebo group and Duration of study drug infusion (h) 1456 (340) 1495 (330)
the percentage of patients who required intraoperative Postoperative analgesia
blood transfusion was less in the dexmedetomidine None 34 (97%) 39 (111%)
group than in the placebo group (table 1 and appendix). Patient-controlled intravenous analgesia 264 (754%) 252 (720%)
After randomisation, a similar proportion of patients Patient-controlled epidural analgesia 52 (149%) 59 (169%)
received supplemental sedation in both groups. However, Use of other analgesics or sedatives during the rst
among patients who received propofol sedation (for 7 postoperative days
mechanical ventilation) after surgery, the total dose of Propofol 178 (509%) 179 (511%)
propofol given was less in the dexmedetomidine group Propofol (mg) 275 (120530) 200 (120400)
than in the placebo group (table 1). Flurbiprofen axetil 110 (314%) 116 (331%)
Postoperative delirium occurred in 79 (23%) of Flurbiprofen axetil (mg) 100 (50100) 100 (625150)
350 patients given placebo, and in 32 (9%) of 350 patients Morphine 102 (291%) 99 (283%)
given dexmedetomidine (odds ratio [OR] 035, 95% CI Morphine (mg) 35 (26) 4 (27)
022054; p<00001), with number needed to treat of Midazolam 34 (97%) 24 (69%)
74 (95% CI 53123) during 7-day follow-up. Per- Midazolam (mg) 25 (161) 4 (210)
protocol analysis also showed a similar dierence in the
Data are number (%), mean (SD), or median (interquartile range). ICU=intensive care unit. APACHE II=Acute Physiology
prevalence of delirium between groups (74 [22%] of
and Chronic Health Evaluation II score. *For prophylaxis of postoperative nausea and vomiting Administered in
334 patients given placebo vs 29 [9%] of 318 patients combination with neostigmine, for reversal of residual neuromuscular blockade. Established with 100 mL of
given dexmedetomidine, OR 035; 95% CI 022056, 05 mg/mL morphine or 125 g/mL sufentanil, programmed to deliver a 2 mL bolus with a lockout interval of
p<00001). Post-hoc analyses showed that daily 610 min and a background infusion of 1 mL/h. Established with 250 mL of 012% ropivacaine plus 05 g/mL
sufentanil, programmed to deliver a 2 mL bolus with a lockout interval of 20 min and a background infusion of 4 mL/h.
prevalence of delirium was signicantly lower in the Dosage among patients who had received the drugs.
dexmedetomidine group than in the placebo group on
postoperative days 13 (day 1: 028, 016050; p<00001, Table 1: Perioperative variables
day 2: 043, 024077; p=0005, day 3: 026, 013053;
p<00001, gure 2). The reduction in the incidence of For patients with endotracheal intubation on ICU
delirium remained when patients were stratied admission, median time to extubation was longer in the
according to the intubation status on ICU admission placebo group than in the dexmedetomidine group
(table 2). Each of the three motoric subtypes of delirium (69 h [95% CI 5286] with placebo and 46 h [3458]
was signicantly decreased in the dexmedetomidine with dexmedetomidine [hazard ratio (HR) 125, 95% CI
group (p<00001). The delirium-sparing eect of low-dose 102153; p=0031]). For all patients, the aggregate
dexmedetomidine became signicant when the duration prevalence of non-delirium complications was reduced
of infusion was 1225 h or longer (table 2). from 73 (21%) of 350 patients given placebo to 52 (15%)

www.thelancet.com Published online August 16, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30580-3 5


of 350 patients given dexmedetomidine (OR 066, drug infusion was modied (infusion rate decreased, or
95% CI 045098; p=0039) (appendix), the length of infusion interrupted temporarily or permanently) were
stay in the ICU was longer in the placebo group, at 215 h signicantly greater (p=0046) in the dexmedetomidine
(207223) than with dexmedetomidine (209 h group than in the placebo group (table 3).
[204214]; HR 118, 102137; p=0027); no signicant
dierences between the two groups were seen in length Discussion
of stay in hospital after surgery and all-cause 30-day Our results suggest that a prophylactic low-dose
mortality. However, in a post-hoc analysis, the percentage dexmedetomidine infusion signicantly decreases the
of patients who were discharged from hospital within incidence of delirium in the rst 7 days after surgery in
7 days after surgery was higher in the dexmedetomidine elderly patients admitted to the ICU after non-cardiac
group (83 [24%] of 350 patients) than in the placebo surgery. This conclusion seems to be true for patients with
group (60 [17%] of 350 patients, OR 150; 104218, or without endotracheal intubation on ICU admission and
p=0032; table 2). for all three motoric subtypes of delirium. Dexmedetomidine
The NRS pain scores both at rest and with movement also signicantly improves the subjective quality of sleep,
were signicantly lower in the dexmedetomidine group decreases the prevalence of non-delirium complications,
than in the placebo group at 3, 6, and 24 h after surgery shortens the length of stay in the ICU, and increases the
(all p<00001, except for one p=0001 at 24 h with percentage of early hospital discharge, but does not
movement); however, the mean dierence of NRS pain signicantly increase adverse events. To our knowledge,
score between groups was small (0 to 1). The NRS ours is the rst suciently powered randomised study that
scores of subjective sleep quality were also signicantly shows the benet of low-dose dexmedetomidine infusion
lower (ie, better) in the dexmedetomidine group than in in this surgical patient population.
the placebo group at 0800 h on the rst, second, and third Postoperative delirium developed in 23% of patients in
days after surgery (all p<00001; table 2). the placebo group, similar to previous studies.1,11,25 In
The RASS scores at the end of study drug infusion keeping with previous reports, the prevalence of
were similar between the two groups. Incidence of postoperative delirium was higher in the patients with
bradycardia and hypotension did not dier between endotracheal intubation on ICU admission than in those
groups, or the percentage of patients requiring without.29 This dierence might be due to a more severe
intervention for these adverse events. On the other hand, underlying condition in intubated patients on ICU
the incidence of tachycardia (p=0002), hypertension admission, which is associated with an increased risk of
(p=0002), and hypoxaemia (p=0001) were signicantly postoperative delirium.30 Secondly, our protocol permitted
lower, and, accordingly, the percentages of patients who the use of supplemental sedatives or analgesics in
required intervention for tachycardia (p=0005) and mechanically ventilated patients during ICU stay;
hypertension (p=0016) were signicantly less in the consequently more of these patients received supplemental
dexmedetomidine group than in the placebo group. propofol, midazolam, and morphine, each of which might
However, the percentages of patients in whom the study increase the risk of postoperative delirium.31
Dexmedetomidine has been used by intensivists in
20 Placebo group general practice for sedation in mechanically ventilated
Dexmedetomidine group ICU patients at an infusion rate from 02 to 17 g/kg per h
with or without a loading dose;79 these sedative doses of
15 p<0001 dexmedetomidine are associated with adverse events,
especially hypotension and bradycardia.79 In this study,
Prevalence of delirium (%)

p=0004 patients were not given a loading dose of

p<0001 dexmedetomidine, and a sub-sedative infusion rate (ie,
10 01 g/kg per h) was given. The RASS scores were similar
between the two groups, indicating that low-dose
dexmedetomidine did not produce signicant sedation.
5 Our protocol was designed to ensure a period of study
drug infusion from ICU admission on the day of surgery
until 0800 h on the rst postoperative morning for a
number of reasons. Firstly, any prophylactic pharma-
1 2 3 4 5 6 7 cological intervention should be initiated early because
Number at risk
Days after surgery the prevalence of delirium is at its highest during the
Placebo group 350 349 346 341 323 310 290 early postoperative hours.30 Secondly, study drug infusion
Dexmedetomidine group 350 349 342 330 307 286 267 should cover the night-time hours, to improve patients
Figure 2: Daily prevalence of postoperative delirium
sleep quality, since dexmedetomidines central action
Sample sizes dier from the rst to seventh day because some patients were discharged from hospital or died converges on the endogenous sleep-promoting pathway.32
during this period. Thirdly, we anticipated that about half of the elective

6 www.thelancet.com Published online August 16, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30580-3


Placebo group (n=350) Dexmedetomidine group OR, HR, or dierence p value

(n=350) (95% CI)
Primary endpoint
Overall incidence of delirium* 79 (226%) 32 (91%) OR=035 (022 to 054) <00001
Secondary endpoints
Time to extubation (h) 69 (52 to 86) (n = 191) 46 (34 to 58) (n = 191) HR=125 (102 to 153) 0031
Overall incidence of non-delirium complications 73 (209%) 52 (149%) OR=066 (045 to 098) 0039
Length of stay in ICU (h) 215 (207 to 223) 209 (204 to 214) HR=118 (102 to 137) 0027
Length of stay in hospital after surgery (day) 110 (102 to 118) 100 (92 to 108) HR=109 (094 to 127) 024
All-cause 30-day mortality 4 (11%) 1 (03%) OR=025 (003 to 223) 021
Prespecied analyses
NRS for pain at rest (score)
3 h after surgery 2 (1 to 4) 2 (0 to 3) D=0 (1 to 0) <00001
6 h after surgery 2 (1 to 3) 1 (0 to 2) D=1 (1 to 0) <00001
24 h after surgery 1 (0 to 3) 1 (0 to 2) D=0 (1 to 0) <00001
NRS for pain with movement (score)
3 h after surgery 3 (2 to 5) 3 (2 to 4) D=1 (1 to 0) <00001
6 h after surgery 3 (2 to 4) 2 (1 to 3) D=1 (1 to 0) <00001
24 h after surgery 2 (1 to 4) 2 (1 to 3) D=0 (1 to 0) 0001
NRS for subjective sleep quality (score)
First morning after surgery 4 (2 to 6) 2 (0 to 4) D=2 (2 to 2) <00001
Second morning after surgery 4 (2 to 6) 2 (1 to 5) D=1 (1 to 1) <00001
Third morning after surgery 4 (2 to 5) 2 (0 to 4) D=1 (2 to 1) <00001
Exploratory analyses
Time to onset of delirium (day) 58 (55 to 60) 65 (64 to 67) HR=039 (026 to 058) <00001
Incidence of delirium according to intubation status on ICU admission
With endotracheal intubation 55 (288%) (n=191) 22 (115%) (n=191) OR=032 (019 to 055) <00001
Without endotracheal intubation 24 (151%) (n=159) 10 (63%) (n=159) OR=038 (017 to 082) 0014
Incidence of delirium according to duration of study drug infusion
<1225 h 17 (183%) (n=93) 11 (143%) (n=82) OR=075 (033 to 170) 049
1225 but <1500 h 17 (250%) (n=68) 7 (103%) (n=68) OR=034 (013 to 090) 0029
1500 but <1758 h 26 (243%) (n=107) 7 (63%) (n=111) OR=021 (009 to 051) 0001
1758 h 19 (232%) (n=82) 7 (74%) (n=94) OR=027 (011 to 067) 0005
Motoric subtype of delirium <00001
None 271 (774%) 318 (909%) .. ..
Hypoactive 42 (120%) 20 (57%) .. ..
Hyperactive 13 (37%) 3 (09%) .. ..
Mixed 24 (69%) 9 (26%) .. ..
Haloperidol treatment 2 (06%) 1 (03%) OR=050 (005 to 552) >099
Time to onset of non-delirium complications (days) 246 (235 to 257) 263 (254 to 273) HR=068 (048 to 098) 0036
Discharge from hospital within 7 days after surgery 60 (171%) 83 (237%) OR=150 (104 to 218) 0032

Data are number (%) or median (95% CI) unless indicated otherwise. OR=odds ratio. HR=hazard ratio. ICU=intensive care unit. NRS=numeric rating scale. D=dierence.
*Occurrence of delirium at any time during the rst 7 days after surgery. Result of patients who were admitted to the ICU with endotracheal intubation. Occurrence of any
non-delirium complications within 30 days after surgery. Data are median (IQR). Stratied according to quartiles of the duration.

Table 2: Eectiveness outcomes

surgical population that require ICU admission would be assigned to receive sedative-inducing doses of
discharged from the ICU within 24 h; by terminating dexmedetomidine versus an alternative sedative-hypnotic
infusion of dexmedetomidine at 0800 h on the rst (benzodiazepines or propofol) or analgesic (opiates) in
postoperative day, enough time has elapsed to enable a mechanically ventilated patients, it was unclear whether
drug with a terminal elimination half-life of 37 h to be those patients beneted through a prophylactic action of
cleared from the plasma by the time the patient is dexmedetomidine, or by avoiding delirium-inducing
discharged from the ICU.33 sedatives and analgesics. Our study was
Although results of previous studies79,31 showed that the placebo-controlled, and a similar proportion of patients
occurrence of delirium was reduced in patients randomly received supplemental sedation in both groups, although

www.thelancet.com Published online August 16, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30580-3 7


Dexmedetomidine can provide analgesia by acting on

Placebo (n=350) Dexmedetomidine p value
(n=350) the 2 adrenergic receptors in the spinal cord.36,37 Our
results also showed that a lower dose of dexmedetomidine
RASS score at the end of study drug infusion 0 (0 to 1) 0 (0 to 2) 012
(scale) signicantly decreased postoperative NRS pain score for
Adverse events up to 24 h after surgery. However, this decrease is small
Bradycardia 46 (131%) 59 (169%) 017
and, therefore, unlikely to have clinical signicance.
Bradycardia with intervention 1 (03%) 5 (14%) 022
Nevertheless, pain itself is a risk factor for the development
Hypotension 92 (263%) 114 (326%) 007
of delirium,38 so a delirium-sparing pharmacoprophylaxis
would seem to be a better strategy than increasing putative
Hypotension with intervention 32 (91%) 34 (97%) 080
delirium-enhancing analgesic drugs to combat
Tachycardia 48 (137%) 23 (66%) 0002
postoperative pain.
Tachycardia with intervention 25 (71%) 9 (26%) 0005
Results of this study show that dexmedetomidine
Hypertension 62 (177%) 34 (97%) 0002
infusion signicantly improves the subjective sleep
Hypertension with intervention 19 (54%) 7 (20%) 0016
quality of postoperative ICU patients, and this benet
Hypoxaemia 50 (143%) 24 (69%) 0001
persists beyond the period of drug infusion, ie, for
Hypoxaemia with intervention 3 (09%) 0 (00%) 025
3 postoperative days, which is consistent with
Modication of study drug infusion* 0046
signicant reduction in daily delirium prevalence.
None 334 (954%) 318 (909%) ..
Dexmedetomidine exerts hypnotic properties by activating
Infusion rate decreased temporarily 1 (03%) 5 (14%) ..
the endogenous sleep-promoting pathway and produces a
Infusion stopped temporarily 14 (40%) 22 (63%) ..
stage II non-rapid eye movement sleep-like state.32 In
Infusion stopped permanently 1 (03%) 5 (14%) .. mechanically ventilated ICU patients, night-time infusion
Data are median (full range) or n (%). RASS=Richmond Agitation Sedation Scale. *Study drug infusion was modied by of dexmedetomidine (0207 g/kg per h) preserved
the attending intensivists before the scheduled end. daynight sleep cycles.39 We speculated that producing a
state akin to natural sleep with dexmedetomidine would
Table 3: Safety outcomes
be a better strategy than delirium-enhancing sedative-
hypnotics, such as benzodiazepines, that produce an
the total consumed propofol dose was less in the patients immobilised state that is unlike natural sleep.40
who received dexmedetomidine (table 1). Additionally, Apart from the dexmedetomidine-induced improvements
post-hoc subgroup analysis showed that dexmedetomidine in hypoxaemia, analgesia, and sleep in this study, a possible
was ecacious in preventing postoperative delirium in mechanism for dexmedetomidines postoperative delirium-
both the intubated and non-intubated patients. Although reducing property might also be attributed to its actions on
the absolute eect size for the delirium-reducing inammation that are evident both clinically41 and in
improvement was higher in the intubated patients, the preclinical models.42 A strong association has been shown
improvement was quite similar (table 2). Importantly, our between elevated biomarkers of inammation and the risk
data also showed that the delirium-sparing eect of for developing delirium;43 the causal nature of this
low-dose dexmedetomidine is dose-dependent, because relationship has been repeatedly shown in preclinical
there is a signicant negative correlation between the studies.44,45 As biomarkers of inammation were not
amount of dexmedetomidine given (infusion duration measured it is not possible to assess whether low-dose
multiplied by rate) and the probability of developing dexmedetomidine infusion suppresses the inammation
delirium (Pearson correlation coecient 0190, associated with the aseptic trauma of surgery.
p<00001). Moreover, dexmedetomidines eect in In respect of safety, our data reveal that
preventing delirium was not limited to the period of drug dexmedetomidine-induced bradycardia and hypotension
infusion but extended up to the third postoperative day were not signicantly increased, possibly because of the
(gure 2). Collectively, taking this investigation together very low doses that were used, whereas hypertension and
with previous studies, one can state with greater tachycardia were both signicantly decreased in patients
condence that the inclusion of dexmedetomidine was given dexmedetomidine. However, our study was
the reason for the delirium-reducing eect. powered for ecacy, but not safety; a larger-scale study
Dexmedetomidine signicantly decreased hypoxaemia will be required to rule out possible safety concerns.
in our patients; results of other studies14,34 have showed Nevertheless, the haemodynamic-stabilising property of
that postoperative hypoxaemia independently contributes dexmedetomidine might be benecial, as this property
to the development of delirium. The underlying reduces adverse cardiac events in high-risk patients.46
mechanism for dexmedetomidines hypoxaemia-sparing Although we did not observe signicant dierences in
action was not directly investigated; preclinical studies cardiovascular events in the present study, the overall
highlight its pulmonary protective eect after remote prevalence of postoperative complications was
organ injury.35 However, this apparent eect could also signicantly lower in the dexmedetomidine group.
just be a coincident signicant nding and warrants In line with previous reports,47 our data showed that
further study. dexmedetomidine shortened the duration of mechanical

8 www.thelancet.com Published online August 16, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30580-3


ventilation and the length of ICU stay, and increased the Declaration of interests
percentage of early hospital discharge. The previously D-XW reports lecture fees and travel expenses for lectures given at
domestic academic meetings from Pzer China, AstraZeneca China,
discussed eects produced by dexmedetomidine Jiangsu Hengrui Medicine Co Ltd, China, and Yichang Humanwell
infusion, including enhanced haemodynamic stability Pharmaceutical Co Ltd, China. MM is supported by National Institutes of
and lowered prevalence of delirium and non-delirium Health R01GM104194, Bethesda, MA, USA. DM is supported by grants
complications, could each contribute to these results.48 from the British Oxygen Chair, and British Journal of Anaesthesia
Fellowship, London, UK. The other authors declare no competing interests.
However, our study does not provide causal relationships Part of the work was presented at the 27th Annual Congress of European
between the various concurrent outcomes. In line with Society of Intensive Care Medicine (ESICM), 2014 in Barcelona, Spain.
our data, a small study49 showed that dexmedetomidine Acknowledgments
decreases time to extubation in mechanically ventilated The authors gratefully acknowledge Xin-Yu Sun (Department of
ICU patients who have agitated delirium. Psychiatrics, Peking University Sixth Hospital, Beijing, China) for her
Although our pragmatic study has many strengths, help in psychiatric consultation. The authors also thank Alex S Evers
(Washington of University School of Medicine, St Louis, USA) for his
including enrolment of a suciently large sample size critical comments during manuscript preparation.
(700 surgical patients) to achieve signicant dierences in
the primary endpoints and some of the secondary 1 Inouye SK, Westendorp RG, Saczynski JS. Delirium in elderly people.
endpoints between the two groups, there are several Lancet 2014; 383: 91122.
limitations. First, the ICU is a noisy and task-intensive 2 Abelha FJ, Luis C, Veiga D, et al. Outcome and quality of life in
patients with postoperative delirium during an ICU stay following
environment that places patients at risk for delirium major surgery. Crit Care 2013; 17: R257.
through sleep deprivation and sleep fragmentation. As we 3 Siddiqi N, House AO, Holmes JD. Occurrence and outcome of
only enrolled ICU surgical patients, it is not certain that delirium in medical in-patients: a systematic literature review.
Age Ageing 2006; 35: 35064.
low-dose dexmedetomidine infusion will lessen sleep 4 Barr J, Fraser GL, Puntillo K, et al. Clinical practice guidelines for
disruptions and thereby decrease delirium in postoperative the management of pain, agitation, and delirium in adult patients
patients in non-ICU settings or medical patients in the in the intensive care unit. Crit Care Med 2013; 41: 263306.
ICU. Second, all participants were screened and enrolled 5 Mo Y, Zimmermann AE. Role of dexmedetomidine for the
prevention and treatment of delirium in intensive care unit
after ICU admission and did not have baseline delirium patients. Ann Pharmacother 2013; 47: 86976.
assessment with cognitive function assessment. 6 Wunsch H, Kahn JM, Kramer AA, et al. Dexmedetomidine in the
Therefore, we cannot preclude the potential bias care of critically ill patients from 2001 to 2007: an observational
cohort study. Anesthesiology 2010; 113: 38694.
introduced by preoperative imbalance of baseline 7 Pandharipande PP, Pun BT, Herr DL, et al. Eect of sedation with
conditions. However, strict randomisation and large dexmedetomidine vs lorazepam on acute brain dysfunction in
sample size should have helped to balance these factors mechanically ventilated patients: the MENDS randomized
controlled trial. JAMA 2007; 298: 264453.
between groups. Third, despite strict randomisation, 8 Xia ZQ, Chen SQ, Yao X, Xie CB, Wen SH, Liu KX. Clinical benets
some baseline and perioperative parameters were not well of dexmedetomidine versus propofol in adult intensive care unit
balanced between the two groups, a not uncommon patients: a meta-analysis of randomized clinical trials. J Surg Res
2013; 185: 83343.
feature in randomised controlled trials. Therefore, we
9 Riker RR, Shehabi Y, Bokesch PM, et al. Dexmedetomidine vs
cannot preclude the potential bias produced by imbalance midazolam for sedation of critically ill patients: a randomized trial.
of these factors. Lastly, it has been noted that CAM-ICU JAMA 2009; 301: 48999.
might not be as sensitive as other tools, eg, 3D-CAM, for 10 Fraser GL, Devlin JW, Worby CP, et al. Benzodiazepine versus
nonbenzodiazepine-based sedation for mechanically ventilated,
delirium assessment especially for other ethnic groups50,51 critically ill adults: a systematic review and meta-analysis of
and in non-ICU settings.52 randomized trials. Crit Care Med 2013; 41: S3038.
Our study indicates that, in elderly patients admitted to 11 Vasilevskis EE, Han JH, Hughes CG, Ely EW. Epidemiology and
risk factors for delirium across hospital settings.
the ICU after non-cardiac surgery, prophylactic low-dose Best Pract Res Clin Anaesthesiol 2012; 26: 27787.
dexmedetomidine infusion signicantly decreases the 12 Bloor BC, Ward DS, Belleville JP, Maze M. Eects of intravenous
prevalence of postoperative delirium. The administration dexmedetomidine in humans. II. Hemodynamic changes.
Anesthesiology 1992; 77: 113442.
of low-dose dexmedetomidine did not signicantly 13 Rosenberg J. Sleep disturbances after non-cardiac surgery.
increase the prevalence of bradycardia or hypotension, Sleep Med Rev 2001; 5: 12937.
but signicantly decreased the prevalence of 14 Flink BJ, Rivelli SK, Cox EA, et al. Obstructive sleep apnea and
hypertension, tachycardia, and hypoxaemia. However, incidence of postoperative delirium after elective knee replacement
in the nondemented elderly. Anesthesiology 2012; 116: 78896.
whether the favourable eects aorded by this novel 15 Alexopoulou C, Kondili E, Diamantaki E, et al. Eects of
application of dexmedetomidine result in improved long- dexmedetomidine on sleep quality in critically ill patients:
term outcomes remains unknown. a pilot study. Anesthesiology 2014; 121: 80107.
16 Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond
Contributors Agitation-Sedation Scale: validity and reliability in adult intensive
D-XW, Z-TM, and DM contributed to the study concept and design. XS, care unit patients. Am J Respir Crit Care Med 2002; 166: 133844.
Z-TM, X-HW, FC, and H-LL contributed to acquisition of data. XS, Z-TM, 17 Inouye SK, Bogardus ST Jr, Williams CS, Leo-Summers L,
D-XW, XZ, and S-NZ analysed and interpreted the data. XS drafted the Agostini JV. The role of adherence on the eectiveness of
manuscript. D-XW, DM, and MM critically revised the manuscript for nonpharmacologic interventions: evidence from the delirium
intellectual content. S-NZ, XS, and D-XW contributed to statistical prevention trial. Arch Intern Med 2003; 163: 95864.
analysis. D-XW supplied administrative, technical, and material support. 18 Inouye SK. Delirium in older persons. N Engl J Med 2006;
Z-TM and D-XW were responsible for study supervision. 354: 115765.

www.thelancet.com Published online August 16, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30580-3 9


19 Cole MG. Delirium in elderly patients. Am J Geriatr Psychiatry 2004; 37 Lin TF, Yeh YC, Lin FS, et al. Eect of combining dexmedetomidine
12: 721. and morphine for intravenous patient-controlled analgesia.
20 Silverstein JH, Timberger M, Reich DL, Uysal S. Central nervous Br J Anaesth 2009; 102: 11722.
system dysfunction after noncardiac surgery and anesthesia in the 38 Kosar CM, Tabloski PA, Travison TG, et al. Eect of preoperative
elderly. Anesthesiology 2007; 106: 62228. pain and depressive symptoms on the development of postoperative
21 Kim HJ, Kim DK, Kim HY, Kim JK, Choi SW. Risk factors of delirium. Lancet Psychiatry 2014; 1: 43136.
emergence agitation in adults undergoing general anesthesia for 39 Oto J, Yamamoto K, Koike S, Onodera M, Imanaka H,
nasal surgery. Clin Exp Otorhinolaryngol 2015; 8: 4651. Nishimura M. Sleep quality of mechanically ventilated patients
22 Munk L, Andersen G, Moller AM. Post-anaesthetic emergence sedated with dexmedetomidine. Intensive Care Med 2012;
delirium in adults: incidence, predictors and consequences. 38: 198289.
Acta Anaesthesiol Scand 2016; published online March 10. 40 Coull JT, Jones ME, Egan TD, Frith CD, Maze M. Attentional eects
DOI:10.1111/aas.12717. of noradrenaline vary with arousal level: selective activation of
23 Ely EW, Inouye SK, Bernard GR, et al. Delirium in mechanically thalamic pulvinar in humans. Neuroimage 2004; 22: 31522.
ventilated patients: validity and reliability of the confusion 41 Memis D, Hekimoglu S, Vatan I, Yandim T, Yuksel M, Sut N.
assessment method for the intensive care unit (CAM-ICU). Eects of midazolam and dexmedetomidine on inammatory
JAMA 2001; 286: 270310. responses and gastric intramucosal pH to sepsis, in critically ill
24 Wang C, Wu Y, Yue P, et al. Delirium assessment using Confusion patients. Br J Anaesth 2007; 98: 55052.
Assessment Method for the Intensive Care Unit in Chinese 42 Xiang H, Hu B, Li Z, Li J. Dexmedetomidine controls systemic
critically ill patients. J Crit Care 2013; 28: 22329. cytokine levels through the cholinergic anti-inammatory pathway.
25 Wang W, Li HL, Wang DX, et al. Haloperidol prophylaxis decreases Inammation 2014; 37: 176370.
delirium incidence in elderly patients after noncardiac surgery: 43 Capri M, Yani SL, Chattat R, et al. Pre-operative, high-IL-6 blood
a randomized controlled trial. Crit Care Med 2012; 40: 73139. level is a risk factor of post-operative delirium onset in old patients.
26 Mu DL, Wang DX, Li LH, et al. High serum cortisol level is Front Endocrinol (Lausanne) 2014; 5: 173.
associated with increased risk of delirium after coronary artery 44 Wan Y, Xu J, Ma D, Zeng Y, Cibelli M, Maze M.
bypass graft surgery: a prospective cohort study. Crit Care 2010; Postoperative impairment of cognitive function in rats: a possible
14: R238. role for cytokine-mediated inammation in the hippocampus.
27 Peterson JF, Pun BT, Dittus RS, et al. Delirium and its motoric Anesthesiology 2007; 106: 43643.
subtypes: a study of 614 critically ill patients. J Am Geriatr Soc 2006; 45 Terrando N, Eriksson LI, Ryu JK, et al. Resolving postoperative
54: 47984. neuroinammation and cognitive decline. Ann Neurol 2011;
28 Ritmala-Castren M, Lakanmaa RL, Virtanen I, Leino-Kilpi H. 70: 98695.
Evaluating adult patients sleep: an integrative literature review in 46 Chalikonda SA. Alpha2-adrenergic agonists and their role in the
critical care. Scand J Caring Sci 2014; 28: 43548. prevention of perioperative adverse cardiac events. AANA J 2009;
29 Burkhart CS, Dell-Kuster S, Gamberini M, et al. Modiable and 77: 10308.
nonmodiable risk factors for postoperative delirium after cardiac 47 Pasin L, Greco T, Feltracco P, et al. Dexmedetomidine as a sedative
surgery with cardiopulmonary bypass. J Cardiothorac Vasc Anesth agent in critically ill patients: a meta-analysis of randomized
2010; 24: 55559. controlled trials. PLoS One 2013; 8: e82913.
30 Shi CM, Wang DX, Chen KS, Gu XE. Incidence and risk factors of 48 Voga G. Hemodynamic changes during weaning: can we assess and
delirium in critically ill patients after non-cardiac surgery. predict cardiac-related weaning failure by transthoracic
Chin Med J (Engl) 2010; 123: 99399. echocardiography? Crit Care 2010; 14: 174.
31 Reade MC, Finfer S. Sedation and delirium in the intensive care 49 Reade MC, Eastwood GM, Bellomo R, et al. Eect of
unit. N Engl J Med 2014; 370: 44454. dexmedetomidine added to standard care on ventilator-free time in
32 Nelson LE, Lu J, Guo T, Saper CB, Franks NP, Maze M. patients with agitated delirium: a randomized clinical trial. JAMA
The alpha2-adrenoceptor agonist dexmedetomidine converges on 2016; 315: 146068.
an endogenous sleep-promoting pathway to exert its sedative 50 Nishimura K, Yokoyama K, Yamauchi N, et al. Sensitivity and
eects. Anesthesiology 2003; 98: 42836. specicity of the Confusion Assessment Method for the Intensive
33 Iirola T, Aantaa R, Laitio R, et al. Pharmacokinetics of prolonged Care Unit (CAM-ICU) and the Intensive Care Delirium Screening
infusion of high-dose dexmedetomidine in critically ill patients. Checklist (ICDSC) for detecting post-cardiac surgery delirium:
Crit Care 2011; 15: R257. a single-center study in Japan. Heart Lung 2016; 45: 1520.
34 Bilotta F, Lauretta MP, Borozdina A, Mizikov VM, Rosa G. 51 van Eijk MM, van den Boogaard M, van Marum RJ, et al.
Postoperative delirium: risk factors, diagnosis and perioperative Routine use of the confusion assessment method for the intensive
care. Minerva Anestesiol 2013; 79: 106676. care unit: a multicenter study. Am J Respir Crit Care Med 2011;
35 Gu J, Chen J, Xia P, Tao G, Zhao H, Ma D. Dexmedetomidine 184: 34044.
attenuates remote lung injury induced by renal ischemia-reperfusion 52 Kuczmarska A, Ngo LH, Guess J, et al. Detection of delirium in
in mice. Acta Anaesthesiol Scand 2011; 55: 127278. hospitalized older general medicine patients: a comparison of the
36 Chrysostomou C, Schmitt CG. Dexmedetomidine: sedation, 3D-CAM and CAM-ICU. J Gen Intern Med 2016; 31: 297303.
analgesia and beyond. Expert Opin Drug Metab Toxicol 2008;
4: 61927.

10 www.thelancet.com Published online August 16, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30580-3