Critically Ill Patients: A Meta-analy

Dexmedetomidine Reduces the Risk of Delirium, Agitation and Confusion in

sis of Randomized Controlled Trials

Laura Pasin, MD, Giovanni Landoni, MD, Pasquale Nardelli, MD, Alessandro Belletti, MD,

Ambra Licia Di Prima, MD, Daiana Taddeo, MD, Francesca Isella, MD, and Alberto Zangrillo, MD

Objectives: Delirium frequently is observed in critically ill

patients in the intensive care unit (ICU) and is associated

strongly with a poor outcome. Dexmedetomidine seems to

reduce time to extubation and ICU stay without detrimental

effects on mortality. The objective of the authors’ study was

to evaluate the effect of this drug on delirium, agitation, and

confusion in the ICU setting.

Design: Meta-analysis of all the randomized clinical trials

ever performed on dexmedetomidine versus any compara-

tor in the ICU setting.

Setting: Intensive care units.

Participants: Critically ill patients.

Interventions: None

Measurements and Main Results: Pertinent studies were

independently searched in BioMedCentral, PubMed,

Embase, and the Cochrane Central Register of clinical trials.

Primary endpoint was the rate of delirium, including the

adverse events, agitation and confusion. The 13 included

manuscripts (14 trials) randomized 3,029 patients. Overall

analysis showed that the use of dexmedetomidine was

Journal of Cardiothoracic and Vascular Anesthesia, Vol ], No ] (Month), 2

associated with significant reductions in the incidence of

delirium, agitation and confusion (298/1,565 [19%] in the

dexmedetomidine group v 337/1,464 [23%] in the control

group, RR ¼ 0.68 [0.49 to 0.96], p ¼ 0.03). Results were

confirmed in subanalyses performed on patients undergoing

noninvasive ventilation (1/53 [2%] in the dexmedetomidine

group v 7/49 [14%] in the control group, RR¼0.18 [0.03 to

1.01], p ¼ 0.05), receiving midazolam as a comparator (268/

1,164 [23%] in the dexmedetomidine group v 277/1,025

[27%] in the control group, RR ¼ 0.68 [0.47 to 1.00], p ¼0.05) and in general ICU setting patients (204/688 [30%] in

the dexmedetomidine group v 204/560 [36%] in the control

group, RR ¼ 0.68 [0.45 to 0.81], p o 0.01).

Conclusions: This meta-analysis of randomized controlled

studies suggests that dexmedetomidine could help to

reduce delirium in critically ill patients.

& 2014 Elsevier Inc. All rights reserved.

KEY WORDS: dexmedetomidine, delirium, agitation,confusion, intensive care unit

From the Department of Anesthesia and Intensive Care, IRCCS SanRaffaele Scientific Institute, Milan, Italy.

Address reprint requests to Giovanni Landoni, MD, Department ofAnesthesia and Intensive Care, San Raffaele Scientific Institute, ViaOlgettina, 60, Milan, Italy. E-mail: landoni.giovanni@hsr.it© 2014 Elsevier Inc. All rights reserved.1053-0770/2601-0001$36.00/0http://dx.doi.org/10.1053/j.jvca.2014.03.010

DELIRIUM IS a commonly observed manifestation inhospitalized patients, with a reported incidence ranging

between 11% and 80% in the intensive care unit (ICU) setting.1

Its pathophysiology is highly heterogenous and not completelyunderstood. One of the primary mechanisms involved seems tobe an imbalance in neurotransmitter systems, in particular areduction of the acetylcholine activity.2 Other involved neuro-transmitter disorders are an excess of serotonin and dopamine,which both affect the cholinergic system.3 Moreover, the rele-ase of gamma-amino-butyricacid (GABA), the most importantinhibitory neurotransmitter in the central nervous system,seems to play an important role in the pathogenesis ofdelirium.4,5 Other hypothesized mechanisms involved arelesions in various areas of the brain (hippocampus, fusiform,or lingual gyri, basal ganglia, thalamus, prefrontal cortex),activation of the inflammatory cascade and an aberrant stressresponse.6 Delirium is associated with prolonged ICU andhospital stay, duration of mechanical ventilation, long-termcognitive impairment and increased mortality after criticalillness.7

Dexmedetomidine is one of the most recently introducedintravenous agents for sedation in the ICU. It is a highlyselective alpha2-adrenergic-receptor agonist, which binds totransmembrane G protein-binding adrenoreceptors in theperiphery (α2A), brain and spinal cord (α2B, α2C) tissues.8

In contrast to other sedative agents, dexmedetomidine, byacting on α2 receptors in the locus coeruleus,9 has analgesiceffects10 without respiratory depression.11,12 All these proper-ties, including a γ-aminobutyric acid receptor-sparing activ-ity,13,14 make dexmedetomidine a potential therapeutic optionfor the management of delirium and agitation in the ICUsetting.

The authors recently published a meta-analysis of allRCTs15 ever performed on dexmedetomidine in mechanically

ventilated patients in the ICU setting suggesting that dexme-detomidine reduces time to extubation and ICU stay. Moreover,the known side effects (increased incidence of bradycardia anda trend toward an increased risk of hypotension) had no effecton overall mortality. The authors, therefore, decided to performan updated meta-analysis of all the randomized clinical trials(RCTs) ever performed on dexmedetomidine versus anycomparator in the ICU setting to evaluate the effect of thisdrug on delirium, agitation, and confusion.

MATERIALS AND METHODS

Search Strategy

Pertinent studies were searched independently in BioMedCentral,PubMed, Embase, and the Cochrane Central Register of clinical trials(updated November 1, 2013) by 4 trained investigators. The fullPubMed search strategy aimed to include any RCTs ever performed inhumans with dexmedetomidine in any clinical setting and is presentedin the supplemental material. In addition, the authors employedbackward snowballing (ie, scanning of references of retrieved articlesand pertinent reviews) and contacted international experts for furtherstudies with no language restriction.

Study Selection

References first were examined independently at a title/abstractlevel by 4 investigators, with divergences resolved by consensus, and

014: pp ]]]–]]] 1

LANDONI ET AL.2

then, if potentially pertinent, retrieved as complete articles. Inclusioncriteria for potentially relevant studies were random allocation totreatment (dexmedetomidine versus any comparator with no restrictionson dose or time of administration) and studies involving ICU patients.The exclusion criteria were duplicate publications, non-adult patientsand lack of data on incidence of delirium, agitation, and/or confusion.(With duplicate publications, in this case, the authors referred to thefirst article published while they retrieved data from the article with thelongest available followup.) Two investigators independently assessedcompliance to selection criteria and selected studies for the finalanalysis with divergences resolved by consensus.

Data Abstraction and Study

Baseline, procedural, and outcome data were abstracted independ-ently by 4 trained investigators (Tables 1 and 2). If a trial reportedmultiple comparisons,16,17 the comparators were aggregated as a singlecontrol group. The primary endpoint of the present review was the rateof delirium, including the adverse events of agitation and/or confusion.In the present meta-analysis, the definition of delirium is modified (notincluding amnesia) from the definition of the Institute of Health andSocial Care (IHSC) as already reported in the paper of Jakob et al18 andincludes all the following: Abnormal behavior, delirium febrile,irritability, affect lability, delirium tremens, mental disorder, aggres-sion, delusion, mental impairment, agitation, depressed level ofconsciousness, mental status changes, altered state of consciousness,depression, metabolic disorder, disorientation, altered mood, neurologicanesthetic complication, disturbance in attention, abnormal neurologicexamination, anger, disturbance in social behavior, psychomotorhyperactivity, dysarthria, restlessness, anxiety, emotional disorder,apathy, emotional distress, screaming, aphasia, euphoric mood, sleepdisorder, brain injury, fear, somnolence, cognitive disorder, hallucina-tion, speech disorder, postoperative confusion, auditory hallucination,abnormal thinking, state of confusion, visual hallucination, impairedjudgment, fluctuating consciousness, hypokinesia, lack of spontaneousspeech, decreased activity, illusion, memory impairment, delirium, andincoherence.

Internal Validity and Risk of Bias Assessment

The internal validity and risk of bias of included trials wereappraised by 2 independent investigators according to the latest versionof the risk of bias assessment tool developed by The CochraneCollaboration,19 with divergences resolved by consensus. Publicationbias was assessed by visually inspecting funnel plots (SupplementalFig 1).

Data Analysis and Synthesis

Computations were performed with Review Manager version 5.2.Hypothesis of statistical heterogeneity was tested by means of CochranQ test with statistical significance set at the two-tailed 0.10 level,whereas extent of statistical consistency was measured with I2, definedas 100% X (Q-df)/Q where Q is Cochran’s heterogeneity statistic anddf the degrees of freedom. Binary outcomes were analyzed to computethe individual and pooled risk ratio (RR) with pertinent 95% confidenceinterval (CI), by means of the same models as just described. Binaryoutcomes from individual studies were analyzed to compute individualand pooled risk ratio (RR) with pertinent 95% confidence interval (CI)by means of an inverse variance method and with a fixed-effect modelin case of low statistical inconsistency (I2 o25%) or with a random-effect model (which better accommodates clinical and statisticalvariations) in case of moderate or high statistical inconsistency(I2 425%). Sensitivity analyses were performed by sequentiallyremoving each study and reanalyzing the remaining dataset (producinga new analysis for each study removed) and by analyzing only datafrom studies with a low risk of bias. Moreover, the authors planned to

perform sensitivity analyses in order to confirm the results of overallanalyses in the subgroups of patients undergoing noninvasive ventila-tion and mechanical ventilation in patients receiving midazolam orpropofol as a comparator drug and in the subgroups of general ICUpatients or elective surgery patients. Statistical significance was set atthe 2-tailed 0.05 level for hypothesis testing. Unadjusted p values arereported throughout. This study was performed in compliance with TheCochrane Collaboration and Preferred Reporting Items for SystematicReviews and Meta-analyses guidelines.20–24

RESULTS

Study Characteristics

Database searches, snowballing, and contacts with expertsyielded a total of 679 articles. The flow chart to select the final13 manuscripts (14 trials)5,16–18,25–33 is detailed in Figure 1.

Excluding 619 nonpertinent titles or abstracts, the authorsretrieved 60 studies in complete form and assessed themaccording to the selection criteria. Forty-seven studies werefurther excluded because of their prespecified exclusion cri-teria. Additional details on excluded studies are available in thesupplemental material.

The 13 included manuscripts randomized 3,029 patients(1,565 to dexmedetomidine and 1,464 to control) (Table 1).Clinical heterogeneity was mostly due to setting, controltreatment, and method of delirium assessment. Indeed, 9 trialsused dexmedetomidine in a general ICU setting5,17,18,27–29,31,33

and 5 in a surgical ICU, either cardiac16,25,26,30 or not.32 Studyquality appraisal indicated that trials were of medium quality(Supplemental Table 1); in particular, 6 of them had a low riskof bias.5,17,18,29,30

Different comparators were identified: Propofol in 4 studyarms,18,25,26,32 midazolam in 5 arms,5,18,27,29,33 placebo in 1arm,28 morphine in 1 arm,30 and propofol plus midazolam in 3arms.16,17,31

Quantitative Data Synthesis

The overall analysis showed that the use of dexmedetomi-dine was associated with significant reductions in the incidenceof delirium, agitation and/or confusion (298/1,565 [19%] inthe dexmedetomidine group v 337/1,464 [23%] in the controlgroup, RR = 0.68 [0.49 to 0.96], p for effect = 0.03, p forheterogeneity o 0.001, I2 = 71% with 13 studies included)(Fig 2) (Table 2).

The significant reduction in delirium was confirmed insubanalyses performed on patients undergoing noninvasiveventilation (1/53 [2%] in the dexmedetomidine group v 7/49[14%] in the control group, RR ¼ 0.18 [0.03 to 1.01], p foreffect ¼ 0.05, p for heterogeneity ¼ 0.94, I2 ¼ 0% with 2studies included) (Fig 3) and patients receiving midazolam asthe comparator drug (268/1164 [23%] in the dexmedetomidinegroup v 277/1,025 [27%] in the control group, RR ¼ 0.68 [0.47to 1.00], p for effect ¼ 0.05, p for heterogeneity o 0.001, I2 ¼76% with 9 studies included) (Fig 4) (Table 2). Moreover,results were confirmed in subanalyses performed in generalICU setting patients (204/688 [30%] in the dexmedetomidinegroup v 204/560 [36%] in the control group, RR ¼ 0.68 [0.45to 0.81], p for effect o 0.01, p for heterogeneity o 0.001, I2 ¼82% with 6 studies included) (Fig 5) (Table 2).

Table 1. Description of the 14 Trials Included in the Meta-analysis

First author Year Setting

Dex

Patients

Control

Patients Dex Loading Dose Dex Infusion Dose Comparator Comparator Dose Delirium Assessment Method

Corbett SM 2005 Cardiac surgery 43 46 1 μg/kg in 15 min 0.4/μg/kg/h Propofol 0.2- 0.7 μg /kg/h or

5-75 μg /kg/min

Adverse event delirium

Herr DL 2003 Cardiac surgery 148 147 1 μg/kg in 20 min 0.2 to 0.7 μg/kg/h Propofol NA Adverse event agitation and

confusion

Huang Z 2012 ICU 33 29 Optional 0.2-0.7 μg/kg/h Midazolam 0.05-0.1 mg/kg/h Adverse event delirium

Jakob SM MIDEX 2012 ICU 249 251 No loading dose 0.2-1.4 μg/kg/h Midazolam 0.03-0.2 mg/kg/h CAM-ICU, adverse events

agitation, confusion

Jakob SM PRODEX 2012 ICU 251 247 No loading dose 0.2-1.4 μg/kg/h Propofol 0.3-4.0 mg/kg/h CAM-ICU, adverse events

agitation, confusion

Maldonado JR 2009 Cardiac surgery 40 38 0,4 μg/kg 0.2-0.7 μg/kg/h Propofol,

midazolam

Propofol: 25-50 μg /kg/min;

Midazolam: 0.5-2 mg/h

DSM-IV-TR

40 40

Martin E 2003 ICU 203 198 1 μg/kg in 10 min 0.2 to 0.7 μg/kg/h Placebo 1 μg /kg for 10 min (loading

dose) and then 0.4 μg /kg/

h. The latter rate could be

adjusted within the range

of 0.2 to 0.7 μg /kg/h

Adverse events agitation and

confusion

Riker RR [33] 2009 ICU 244 122 1μg/kg 0.2-1.4 μg/kg/h Midazolam Loading dose 0.05 mg/kg

then infusion rate 0.02-0.1

mg/kg/h

CAM-ICU

Ruokonen E 2009 ICU 41 44 No loading dose 0.8 μg/kg/h for 1

h and then

adjusted

stepwise at

0.25, 0.5, 0.8,

1.1, and 1.4

μg/kg/h

Propofol

Midazolam

Propofol: 2.4 mg/kg/h for 1 h

and then adjusted

stepwise at 0.8, 1.6, 2.4,

3.2, and 4.0 mg/kg/h;

Midazolam: boluses (1-2

mg), starting at 3 boluses

per hour for 1 h, and

thereafter 1-4 boluses per

h, and if not sufficient as

continuous infusion

CAM-ICU, adverse event

confusion

Senoglu N 2010 ICU 20 20 1 μg/kg over

10 minutes

0.5 μg/kg/h Midazolam Loading dose of 0.05 mg/kg

midazolam over 10

minutes followed by a

maintenance infusion of

0.1 mg/kg/h

Adverse event agitation

Shehabi Y 2009 Cardiac surgery 154 152 No loading dose 0.1-0.7 μg/kg/mL Morphine 10-70 μg /kg/mL CAM-ICU

Shehabi Y 2013 ICU 21 16 No loading dose 1 mg/kg/h Propofol,

Midazolam

At the discretion of the

treating clinician

CAM-ICU

Terao Y 2012 Cervical spine

surgery

16 16 0.1 μg/kg/min in

10 minutes

0.4 μg/kg/h Propofol 0.1 mg/kg/min for 10 minutes

as a loading dose,

followed by a continuous

infusion at 1 mg/kg/h

Adverse event agitation

Wan LJ 2011 ICU 102 98 NA Midazolam NA Adverse event delirium

Abbreviations: CAM-ICU, Confusion Assessment Method for the ICU; Dex, dexmedetomidine; DSM-IV-TR, diagnostic and statistical manual of mental disorders; ICU, Intensive Care Unit; NA, not

available.

DEXMEDETOMID

INEAND

DELIRIU

M3

Table 2. Primary Outcome and Sensitivity Analyses

Outcome

Number of

Included Trials Dex Patients Control Patients RR 95% CI P for Effect

P for

Heterogeneity I2 (%)

Overall trials 14 trials

(13 Manuscripts)

1,565 1,464

Delirium, agitation and/or confusion 14 298/1,565 [19%] 337/1,464 [23%] 0.68 0.49 to 0.96 0.03 o 0.001 71

Midazolam as comparator drug 6 204/688 [30%] 204/360 [36%] 0.45 0.45 to 0.81 o 0.01 o 0.001 82

Propofol as comparator drug 5 54/498 [11%] 82/494 [17%] 0.79 0.32 to 1.96 0.61 0.02 65

General ICU patients 9 268/1164 [23%] 277/1025 [27%] 0.68 0.47 to 1.00 0.05 o 0.001 76

Elective surgery patients 5 30/401 [7%] 60/439 [14%] 0.80 0.30 to 2.16 0.66 0.02 66

Mechanically ventilated patients 12 297/1512 [20%] 330/1415 [23%] 0.71 0.51 to 1.01 0.06 o 0.001 74

Patients on NIV 2 1/53 [2%] 7/49 [14%] 0.18 0.03 to 1.01 0.05 0.94 0

Delirium defined by CAM-ICU or

DSM-IV-TR monitoring tools

4 157/459 [34%] 154/368 [42%] 0.61 0.39 to 0.96 0.03 0.08 56

SENSITIVITY ANALYSIS (including

only low risk of bias studies)

6 263/959[27%] 245/836 [29%] 0.83 0.60 to 1.15 0.26 0.004 71

SENSITIVITY ANALYSIS (removing

1 study at time)

All 95% CIs of RR 41 and p o 0.05

NOTE. The significant reduction in delirium was confirmed only in subanalyses performed on patients undergoing noninvasive ventilation, in

patients receiving midazolam as a comparator drug and in general ICU setting patients. Results were not confirmed when sensitivity analyses

including only low risk of bias studies were performed.

Abbreviations: CAM-ICU, Confusion Assessment Method for the ICU; CI, confidence interval; Dex, dexmedetomidine; DSM-IV-TR, diagnostic

and statistical manual of mental disorders; NIV, noninvasive ventilation; P, p value; RR, relative risk.

LANDONI ET AL.4

The different screening tools used to assess delirium did notaffect results (Table 1).

Visual inspection of the funnel plot did not identify askewed or asymmetric shape, excluding the presence of smallpublication bias (Supplemental Material).

Results were not confirmed when sensitivity analysesincluding only low risk of bias studies were performed inorder to determine the “robustness” of the results (Table 2).

DISCUSSION

The authors’ meta-analysis suggests that dexmedetomidinereduces the risk of delirium, agitation, and/or confusion in

Fig 1. Flow chart to select t

critically ill patients. The findings were confirmed in severalsubsettings such as patients undergoing noninvasive mechan-ical ventilation and general/medical ICU patients.

Several studies tried to demonstrate that the favorablepharmacologic characteristics of dexmedetomidine could helpto prevent or reduce the incidence of delirium in critically illpatients. The results of these studies, although encouraging,remained controversial.

Two theories have been proposed by Maldonado et al16 toexplain the possible mechanisms for a reduction in theincidence of delirium with the use of dexmedetomidine. Thefirst one hypothesizes an intrinsic delirium-saving property of

he final 13 manuscripts.

Fig 2. Forest plot for incidence of delirium, agitation and/or confusion in the overall population.

DEXMEDETOMIDINE AND DELIRIUM 5

dexmedetomidine, determined by the multiple characteristics ofthe drug. In fact, like all α2-adrenoceptor agonists, it hasminimal effects on cognitive impairment. On the contrary,GABAergic drugs may contribute to the development ofdelirium since GABA, the primary inhibitory neurotransmitterin the central nervous system, seems to be involved strongly inthe pathogenesis of delirium. Secondly, dexmedetomidine mayhave a positive effect on reducing delirium since it decreasesthe need for GABAergic agents, benzodiazepines andopioids.15 Moreover, it does not have significant effects onthe cholinergic system, one of the most involved in thecognitive functions and in the subsequent development ofdelirium.

Tan and Ho,34 in the first meta-analysis of RCTs everperformed on dexmedetomidine in critically ill patients, failedto demonstrate the superiority of dexmedetomidine in reducingthe risk of delirium when compared with other sedative drugs.

The American College of Critical Care Medicine, in therecent revised “Clinical Practice Guidelines for the SustainedUse of Sedatives and Analgesics in the Critically Ill Adult,”35

stated that in mechanically ventilated patients at risk ofdeveloping delirium in the ICU, dexmedetomidine infusionmay be associated with a lower prevalence of deliriumcompared to benzodiazepine infusion. On the other hand, theyprovide no recommendation for the use of dexmedetomidine toprevent delirium, as there is no conclusive evidence regardingits effectiveness in these patients. Moreover, they provide no

Fig 3. Forest plot for incidence of delirium, agitation, and/or

recommendation for using either a pharmacologic or a com-bined nonpharmacologic and pharmacologic delirium preven-tion protocol. They do not suggest the use of either haloperidolor atypical antipsychotics but recommend performing earlymobilization of patients in order to reduce the incidence andduration of delirium.

Concerning delirium treatment, the American College ofCritical Care Medicine states that atypical antipsychotics mayreduce the duration of delirium and suggest using an analgesia-first sedation. They do not recommend administering rivastig-mine or antipsychotics (in patients at significant risk fortorsades de pointes), but recommend either daily sedationinterruption or a light target level of sedation in mechanicallyventilated patients and promoting sleep.

Fraser et al,36 in a subsequent meta-analysis of RCTscomparing benzodiazepine versus non-benzodiazepine-basedsedation for mechanically ventilated patients in the ICU setting,confirmed that data on delirium prevalence were insufficient todraw clear conclusions.

Mo and Zimmermann,37 in a recent narrative reviewfocusing on the role of dexmedetomidine for prevention andtreatment of delirium in ICU patients, showed that currentlyavailable evidence suggests that dexmedetomidine is a promis-ing agent not only for prevention but also for treatment ofdelirium in critically ill patients. However, the authors statedthat the role of dexmedetomidine in preventing delirium neededto be further clarified.

confusion in patients undergoing noninvasive ventilation.

Fig 4. Forest plot for incidence of delirium, agitation, and/or confusion (dexmedetomidine versus midazolam).

LANDONI ET AL.6

On the other hand, a recent meta-analysis38 of RCTs showedthat dexmedetomidine may decrease the risk of delirium whencompared to propofol as a sedative agent in ICU patients.

The authors’ meta-analysis differs from all the above-citedstudies since it is the first one focusing only on delirium andincluding all RCTs ever performed on dexmedetomidine versusany comparator in the ICU setting (not only propofol).

The authors recently published a meta-analysis on dexme-detomidine versus any comparator in mechanically ventilatedpatients15 showing that the use of dexmedetomidine wasassociated with reductions in the length of ICU stay (weightedmean difference (WMD) ¼ �0.79 [�1.17 to �0.40] days, pfor effect o 0.001), in time to extubation (WMD ¼ �2.74[�3.80 to �1.65] hours, p for effect o0.001) and in the needfor additional doses of both sedative (propofol, benzodiaze-pines, or antipsychotics) or analgesic drugs (opioids). More-over, the authors found that the known side effects ofdexmedetomidine (bradycardia and hypotension) had no effecton overall mortality. The present meta-analysis differed fromthe above-cited one since it focused on delirium (that was notinvestigated previously), was more updated, and it was notlimited to mechanically ventilated patients but included allstudies performed in the ICU setting.

A strength of the authors’ meta-analysis was that, asdescribed in details in the methods section, they chose asprimary end-point of their study the composite end-pointdelirium, agitation and/or confusion. In fact, although deliriumis a well-defined clinical condition, many different terms are

Fig 5. Forest plot for incidence of delirium, agita

used to describe it in medical literature.39 According to therecently released diagnostic criteria of the Diagnostic andStatistical Manual or DSM-540 and the last edition of theInternational Classification of Mental and Behavioural Disordersor ICD-10,41 delirium can be defined as a syndrome of abruptonset and fluctuating course with marked cognitive symptomsincluding decreased attention and awareness, cognitive impair-ment, and optional additional deficits such as memory, disori-entation, or psychomotor changes. Different standardizedmonitoring tools, such as the Confusion Assessment Methodfor the ICU (CAM-ICU), the Diagnostic and Statistical Manualof Mental Disorders (DSM-IV) or the Intensive Care DeliriumScreening Checklist (ICDSC), commonly are used to detectdelirium in non-verbal, mechanically ventilated patients. Itshould not be underestimated, however, that physicians withoutpsychiatric training often report adverse events such as agitation,disorientation, or confusion without classifying them as deliriumwith a standardized monitoring tool. The above considerationcould significantly affect the evaluation of the outcome incritically ill patients. The authors, therefore, decided to includeall these definitions in the definition of their primary endpointand to perform different subgroup analyses accordingly with thetype of screening tool used.

The pharmacokinetic properties of fast bioelimination canexplain why dexmedetomidine use results in a shorter time toextubation when compared with other sedatives or hypnoticagents.10 Since delirium, agitation, and confusion after sedationare frequent in critically ill patients, it is reasonable to think that

tion, and/or confusion in general ICU setting.

DEXMEDETOMIDINE AND DELIRIUM 7

dexmedetomidine might reduce time to extubation and length ofICU stay through a reduction in the delirium rate. Anotherpossible mechanism of action is represented by the α2-adrenergic-receptor agonist effect in the central nervous system.

The authors’ meta-analysis presents some limitations. First ofall, the included studies were of medium-low quality. Secondly,they found high heterogeneity among groups, both in the overallpopulation and in the different subgroups of analyzed patients.Thirdly, the findings were not confirmed when including onlylow-risk-of-bias studies, suggesting that there still is need tofurther investigate this topic with high-quality trials.

The use of dexmedetomidine seems to reduce the incidenceof delirium in critically ill patients, in particular in the generalICU setting and in the subgroup of patients requiring non-invasive mechanical ventilation. These advantages are partic-ularly encouraging when dexmedetomidine is compared withmidazolam. Large, multicenter, randomized clinical trialswould be welcome to confirm these findings.

APPENDIX A. SUPPLEMENTARY DATA

Supplementary data associated with this article can be foundin the online version at doi:10.1053/j.jvca.2014.03.010.

REFERENCES

1. Ouimet S, Kavanagh BP, Gottfried SB, et al: Incidence, riskfactors and consequences of ICU delirium. Intensive Care Med 33:66-73, 20072. Hshieh TT, Fong TG, Marcantonio ER, et al: Cholinergic

deficiency hypothesis in delirium: A synthesis of current evidence.J Gerontol A Biol Sci Med Sci 63:764-772, 20083. Trzepacz PT: Update on the neuropathogenesis of delirium.

Dement Geriatr Cogn Disord 10:330-334, 19994. Pandharipande P, Shintani A, Peterson J, et al: Lorazepam is an

independent risk factor for transitioning to delirium in intensive careunit patients. Anesthesiology 104:21-26, 20065. Riker RR, Shehabi Y, Bokesch PM, et al: Dexmedetomidine vs

midazolam for sedation of critically ill patients: A randomized trial.JAMA 301:489-499, 20096. Zaal IJ1 Slooter AJ: Delirium in critically ill patients: Epidemi-

ology, pathophysiology, diagnosis and management. Drugs 72(1):457-1,471, 20127. Zhang Z, Pan L, Ni H: Impact of delirium on clinical outcome in

critically ill patients: A meta-analysis. Gen Hosp Psychiatry 35:105-111,20138. Paris A, Tonner PH: Dexmedetomidine in anaesthesia. Curr Opin

Anaesthesiol 18:412-418, 20059. Huupponen E, Maksimow A, Lapinlampi P, et al: Electroence-

phalogram spindle activity during dexmedetomidine sedation andphysiological sleep. Acta Anaesthesiol Scand 52:289-294, 200810. Guo TZ, Jiang JY, Buttermann AE, et al: Dexmedetomidine

injection into the locus ceruleus produces antinociception. Anesthesi-ology 84:873-881, 199611. Iirola T, Aantaa R, Laitio R, et al: Pharmacokinetics of prolonged

infusion of high-dose dexmedetomidine in critically ill patients. CritCare 15:R257, 201112. Morsch R, Ferri M, Vasconcelos C, et al: Dexmedetomidine as a

sedative agent for more than 24 hours in acutely ill patients. Crit Care9:P111, 200513. Gerlach AT, Murphy CV, Dasta JF: An updated focused review

of dexmedetomidine in adults. Ann Pharmacother 43:2064-2074, 200914. Yu SB: Dexmedetomidine sedation in ICU. Korean J Anesthesiol

62:405-411, 201215. Pasin L, Greco T, Feltracco P, et al: Dexmedetomidine as a

sedative agent in critically ill patients: A meta-analysis of randomizedcontrolled trials. PLoS One 8:e82913, 201316. Maldonado JR, Wysong A, van der Starre PJ, et al: Dexmede-

tomidine and the reduction of postoperative delirium after cardiacsurgery. Psychosomatics 50:206-217, 200917. Ruokonen E, Parviainen I, Jakob SM, et al: Dexmedetomidine

for Continuous Sedation I. Dexmedetomidine versus propofol/midazo-lam for long-term sedation during mechanical ventilation. IntensiveCare Med 35:282-290, 2009

18. Jakob SM, Ruokonen E, Grounds RM, et al: Dexmedetomidinefor Long-Term Sedation I. Dexmedetomidine vs midazolam orpropofol for sedation during prolonged mechanical ventilation: Tworandomized controlled trials. JAMA 307:1151-1160, 201219. Higgins JPT, Altman DG, Sterne JAC (eds). Chapter 8: Assess-

ing risk of bias in included studies. Cochrane Handbook for SystematicReviews of Interventions. Version 5.1.0 [updated March 2011]. TheCochrane Collaboration, 2011. Available from www.cochrane-handbook.org.20. Begg CB, Mazumdar M: Operating characteristics of a rank

correlation test for publication bias. Biometrics 50:1088-1101, 199421. Peters JL, Sutton AJ, Jones DR, et al: Comparison of two methods

to detect publication bias in meta-analysis. JAMA 295:676-680, 200622. Biondi-Zoccai G, Lotrionte M, Landoni G, et al: The rough guide

to systematic reviews and meta-analyses. HSR Proc Intensive CareCardiovasc Anesth 3:161-173, 201123. Liberati A, Altman DG, Tetzlaff J, et al: The PRISMA statement

for reporting systematic reviews and meta-analyses of studies thatevaluate healthcare interventions: Explanation and elaboration. BMJ339:b2700, 200924. Greco T, Zangrillo A, Biondi-Zoccal G, et al: Meta-analysis:

Pitfalls and hints. Heart Lung Vessel 5:219-225, 201325. Corbett SM, Rebuck JA, Greene CM, et al: Dexmedetomidine

does not improve patient satisfaction when compared with propofolduring mechanical ventilation. Crit Care Med 33:940-945, 200526. Herr DL, Sum-Ping ST, England M: ICU sedation after coronary

artery bypass graft surgery: Dexmedetomidine-based versus propofol-based sedation regimens. J Cardiothorac Vasc Anesth 17:576-584, 200327. Huang Z, Chen YS, Yang ZL, et al: Dexmedetomidine versus

midazolam for the sedation of patients with non-invasive ventilationfailure. Intern Med 51:2299-2305, 201228. Martin E, Ramsay G, Mantz J, et al: The role of the alpha2-

adrenoceptor agonist dexmedetomidine in postsurgical sedation in theintensive care unit. J Intensive Care Med 18:29-41, 200329. Senoglu N, Oksuz H, Dogan Z, et al: Sedation during non-

invasive mechanical ventilation with dexmedetomidine or midazolam:A randomized, double-blind, prospective study. Curr Ther Res ClinExp 71:141-153, 201030. Shehabi Y, Grant P, Wolfenden H, et al: Prevalence of delirium

with dexmedetomidine compared with morphine based therapy aftercardiac surgery: A randomized controlled trial (DEXmedetomidineCOmpared to Morphine-DEXCOM Study). Anesthesiology 111:1075-1084, 200931. Shehabi Y, Bellomo R, Reade MC, et al: Sedation Practice in

Intensive Care Evaluation Study I, Australian, New Zealand IntensiveCare Society Clinical Trials G. Early goal-directed sedation versusstandard sedation in mechanically ventilated critically ill patients: Apilot study. Crit Care Med 41:1983-1991, 2013

LANDONI ET AL.8

32. Terao Y, Ichinomiya T, Higashijima U, et al: Comparisonbetween propofol and dexmedetomidine in postoperative sedation afterextensive cervical spine surgery. J Anesth 26:179-186, 201233. Wan LJ, Huang QQ, Yue JX, et al: [Comparison of sedative

effect of dexmedetomidine and midazolam for post-operativepatients undergoing mechanical ventilation in surgical inten-sive care unit]. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 23:543-546, 201134. Tan JA, Ho KM: Use of dexmedetomidine as a sedative and

analgesic agent in critically ill adult patients: A meta-analysis. IntensiveCare Med 36:926-939, 201035. Barr J, Fraser GL, Puntillo K, et al: Clinical practice

guidelines for the management of pain, agitation, and delirium inadult patients in the intensive care unit. Crit Care Med 41:263-306,201336. Fraser GL, Devlin JW, Worby CP, et al: Benzodiazepine versus

nonbenzodiazepine-based sedation for mechanically ventilated,

critically ill adults: A systematic review and meta-analysis of random-ized trials. Crit Care Med 41:S30-S38, 201337. Mo Y, Zimmermann AE: Role of dexmedetomidine for the

prevention and treatment of delirium in intensive care unit patients.Ann Pharmacother 47:869-876, 201338. Xia ZQ, Chen SQ, Yao X, et al: Clinical benefits of dexmede-

tomidine versus propofol in adult intensive care unit patients: A meta-analysis of randomized clinical trials. J Surg Res 185:833-843, 201339. Morandi A, Pandharipande P, Trabucchi M, et al: Understanding

international differences in terminology for delirium and other types ofacute brain dysfunction in critically ill patients. Intensive Care Med 34:1907-1915, 200840. American Psychiatric Association DSM 5 task force: Diagnostic

and statistical manual of mental disorders: DSM-5, 2013.41. World Health Organization: International Statistical Classifica-

tion of Diseases and Related Health Problems 10th Revision (ICD-10)Version for 2010, 2011.