Patterns of Postoperative Delirium in Children*
Jochen Meyburg, MD1; Mona-Lisa Dill1; Chani Traube, MD2; Gabrielle Silver, MD3;
Rebecca von Haken, MD4
Objective: Intensive care delirium is a substantial problem in
adults. Intensive care delirium is increasingly recognized in pedi-
atrics in parallel with the development of specific scoring systems
for children. However, little is known about the fluctuating course
of intensive care delirium in children after surgery and possible
implications on diagnostic and therapeutic strategies.
Design: Patients that needed treatment in the PICU following
elective surgery were screened for intensive care delirium with
the Cornell Assessment of Pediatric Delirium. When the patients
were awake (Richmond Agitation and Sedation Score > -3), two
trained investigators conducted the Cornell Assessment of Pedi-
atric Delirium twice daily for five consecutive days.
Patients: Ninety-three patients aged 0 to 17 years.
Interventions: Eight hundred forty-five assessments completed.
Measurements and Main Results: Of the 845 scores, 230 were
consistent with delirium (27.2%). Sixty-one patients (65.5%) were
diagnosed with intensive care delirium. Half of these patients
(n = 30; 32.2%) had a short-lasting delirium that resolved within
24 hours, and half (n = 31; 33.3%) had delirium of longer dura-
tion. Delirium could be clearly distinguished from sedation by anal-
ysis of individual test items of the Cornell Assessment of Pediatric
Delirium. Time spent delirious had a measurable effect on out-
come variables, including hospital length of stay.
Conclusion: Most postoperative PICU patients develop intensive
care delirium. Some have a short-lasting course, which underlines the
*See also p. 191.
1
Department of General Pediatrics, University Children’s Hospital,
Heidelberg, Germany.
2
Pediatric Critical Care Medicine, Weill Cornell Medical College, New
York, NY.
3
Department of Child and Adolescent Psychiatry, Weill Cornell Medical
College, New York, NY.
4
Department of Anesthesiology, University Hospital Heidelberg, Heidel-
berg, Germany.
Drs. Traube and Silver received support for travel from Weill Cornell Medi-
cal College. Dr. Meyburg received support for travel from Heidelberg Uni-
versity Children’s Hospital. Dr. von Haken received funding from lectures
on delirium in adults, and from Orionpharma; her institution received fund-
ing from Köhler Chemie. The remaining authors have disclosed that they
do not have any potential conflicts of interest.
For information regarding this article, E-mail: jochen.meyburg@med.uni-
heidelberg.de
Copyright © 2017 by the Society of Critical Care Medicine and the World
Federation of Pediatric Intensive and Critical Care Societies
DOI: 10.1097/PCC.0000000000000993
128 www.pccmjournal.org February 2017 • Volume 18 • Number 2
Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
need for early screening. Our findings support the view of delirium
as a continuum of acute neurocognitive disorder. Further research
is needed to investigate prophylactic and treatment approaches for
intensive care delirium. (Pediatr Crit Care Med 2017; 18:128–133)
Key Words: Cornell Assessment of Pediatric Delirium; intensive
care delirium; pediatric delirium; subsyndromal delirium
D
elirium as an acute brain dysfunction has become an
important issue in adult critical care, since affected
patients have a higher mortality (1) and a significant
risk of long-term neurocognitive impairment (2). Over the
past years, there has been rising awareness that intensive care
delirium (ICD) is not restricted to adult patients. First reports
from the 1980s and 1990s described behavioral disturbances
in children on intensive care wards but did not classify them as
ICD (3, 4). A prospective pediatric study by Schieveld et al (5)
found a very low prevalence of ICD of only 4.5% among psychiat-
ric referrals. With the adaptation of delirium scores to the special
demands in children, higher rates of delirium up to 16.9% were rec-
ognized (6). Recent studies using scoring systems that are suitable
for children under 2 years old, that is, the Cornell Assessment of
Pediatric Delirium (CAPD) and the Preschool Confusion Assess-
ment Method for the ICU (psCAM-ICU), yielded overall delirium
rates of 21–44% (7, 8). The true rate of ICD in children may be even
higher, since up to 82% of ventilated adult patients are affected (9).
Because recognition of pediatric delirium is fairly novel, there
has been little opportunity to describe the course of delirium
in critically ill children. One of the key features of delirium is
its fluctuating course, which may have significant implications
on screening algorithms. Two retrospective studies (10, 11) only
scored their patients once. In the early prospective studies by
the Dutch group, patients were also assessed once, followed by
a neuropsychologic investigation on the same day (5, 6, 12). The
CAPD and the Pediatric Confusion Assessment Method for the
ICU (pCAM-ICU) are quick bedside tools and thus can be uti-
lized repeatedly in individual patients. The most recent studies
have scored at least some of their patients more than once, but
no details on the course of the score results are given (7, 8, 13).
The aim of this study was to learn more about the time course
of ICD in children after surgery. There are two reasons why this
is important. First, incomplete presentations of ICD are well
 Feature Articles

known in adult patients as subsyndromal delirium (SSD). SSD
can progress to full-blown ICD, and it is discussed whether such
deterioration can be prevented by early recognition and treat-
ment. Little if anything is known about partial or subsyndromal
ICD in children. Second, there are no generally accepted guide-
lines for the treatment of pediatric ICD. Given the potential harm
caused by psychotropic medications (14), it is imperative that we
learn more about the duration and resolution of symptoms in
affected patients. A better understanding of the distinct features
of pediatric delirium, and how they emerge and evolve, may help
in better defining severity and impact of ICD.
METHODS
Study Design
This study was conducted in a 22-bed general PICU in a tertiary-
care academic medical center over a 6-month period from May
2014 to October 2014. All patients admitted to the PICU follow-
ing elective surgery were included. The only exclusion criterion
was refusal by the parents to provide informed consent prior to
the operation. When appropriate, assent was obtained from the
child as well. Demographic and clinical data were collected for all
study subjects. The study was approved by the Institutional Review
Board of the University Hospital Heidelberg.
respectively. Chi-square test was used for discrete variables. Delir-
ium scores were analyzed by the total CAPD score, and by indi-
vidual test-item scores. Subgroup analysis of patients with delirium
was analyzed by analysis of variance. Analysis of covariance was
used to assess the effect of delirium on hospital length of stay (LOS)
after adjusting for disease severity and time to first assessment. p
values of less than 0.05 were regarded as statistically significant. All
analyses were performed in SPSS 20 (IBM, Armonk, NY).
RESULTS
Demographic Data
Demographic data are given in Table 1. A total of 93 children
were included in the study, of which 61 (65.6%) were delirious
in at least one assessment (CAPD, ≥ 9). The sex ratio was equal in
the two groups of patients with (male:female, 31:30) and with-
out delirium (16:16), respectively. Patients with delirium were
significantly younger compared to the control group (2.0 ± 3.2 vs
6.4 ± 5.7 yr; p < 0.001). It was not possible to analyze subgroups
Demographic Details and
Table 1.
Admission Diagnoses of Subjects
Characteristic n (%)

Patients included 93
Assessment of Postoperative Delirium, Severity of
Male 47
Illness, and Level of Care
The authorized German version of the CAPD (15) was used for Female 46
the assessment of postoperative delirium. The CAPD is an obser- Patient age, yr, n (%)
vational screen with eight individual test items that correlate with
 0–1 44 (47.3)
the Diagnostic and Statistical Manual-V diagnostic criteria for
delirium (16). Each item is scored 0–4. A total score of greater than  1–6 30 (32.3)
or equal to 9 is consistent with a diagnosis of delirium. Screen-  6–12 9 (9.7)
ing was started when the Richmond Agitation and Sedation
Score (RASS) was greater than –3, that is, the child would briefly  12–18 10 (10.8)
awaken to voice, open his eyes, and make short eye contact (17). Diagnoses
All assessments were done by the same two trained investigators   Congenital heart disease 58
together with the nurse in charge of the patient. Scoring was done
twice daily in the middle of the nurses’ morning and at the end of   CNS tumor or malformation 7
the evening shifts, respectively. Over a period of five consecutive   Urologic malformation 5
days, a total of 10 assessments per patient were made. Scoring   Tethered cord 5
was continued in the step-down unit if a patient was transferred
from the PICU before the end of the assessment period. To  Craniosynostosis 5
determine severity of illness, the revised version of the Pediatric   Chronic renal failure requiring 5
Index of Mortality 2 (PIM2) was calculated at the timepoint of transplantation
the first assessment (18). The level of care was determined using   Airway malformations 3
the Intensive Nursing and Performance System classification, a
 Meningomyelocele 2
validated system that is used in several German ICUs. Based on a
detailed scoring system, patients are assigned once daily to differ-   Biliary atresia 2
ent categories from 1 (intermediate care patient, spontaneously   Oesophageal atresia 1
breathing) to 6 (patient with multiple organ failure requiring
more than one extracorporeal support system) (19). Mechanical ventilation on the PICU, n (%)
 None 23 (24.7)
Statistical Analysis   < 24 hr 42 (45.2)
After testing for normal distribution, Student t test was used to
  ≥ 24 hr 28 (30.1)
compare the two groups of patients with and without delirium,

Pediatric Critical Care Medicine www.pccmjournal.org 129

Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
 Meyburg et al

of patients based on their ethnicity, since the families originated
from a total of 25 countries. As a substitute with regard to pos-
sible sociocultural differences, we grouped patients as having a
migrant background or not, as defined by German authorities.
However, no significant differences were found between the two
clinical groups. There were no significant differences in disease
severity (PIM2) between delirious children and children who
never developed clinical symptoms of delirium. Type of surgery,
use of cardiac bypass, or the extent of hyperthermia or blood
loss did not have an impact on the development of delirium.
Delirium Patterns
Interestingly, two different patterns of ICD were found in our
postoperative patients. Half of the delirious patients (30/61,
group 1) had an early short-lasting delirium (24 hr), and
the other half (31/61, group 2) had delirium of longer dura-
tion, with higher scores (12.5 ± 4.0 vs 15.8 ± 5.0 [mean ± sd];
p < 0.001) throughout the study period (Fig. 1). Eleven of these
more severely affected children remained delirious throughout
the 5-day study period.
To test whether the mildly elevated scores in group 1 were
sedation falsely classified as delirium, further analyses were
made. Each item in the CAPD was analyzed separately. The
analysis of the individual test items is shown in Figure 2. Each
of the eight items of the CAPD can be rated with 0–4 points. In
the children who did not develop delirium (group 0; Fig 2A),
mean values for all test items were close to zero. In children
with delirium greater than 24 hours (group 2; Fig 2B), mean
values were significantly higher, with markedly elevated values
for items 2, 5, and 8. In children with short-lasting delirium
(group 1), normal and pathologic scores were analyzed sepa-
rately. We found that during asymptomatic periods (Fig 2C),
the values were almost identical to unaffected children (Fig
2A). In contrast, during the short periods of delirium (Fig 2D),
the pattern of the test items closely resembled the distribution
in the severely affected group 2 patients (pattern B). Table 2

Figure 1. Cornell Assessment of Pediatric Delirium (CAPD) scores (mean and sd) in children without delirium (white bars), with mild delirium (gray bars),
and severe delirium (black bars). Patients were investigated twice daily for five consecutive days (investigations 1–10) starting at a Richmond Agitation and
Sedation Score greater than –3. Compared to patients without delirium: **p < 0.01 and *p < 0.05. Compared to patients with mild delirium: #p < 0.01.

130 www.pccmjournal.org February 2017 • Volume 18 • Number 2

Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
 Feature Articles

A B the three groups. Compared to

patients who were never deliri-
ous, patients with longer lasting
delirium required longer periods
of mechanical ventilation, took
longer to emerge from sedation,
had longer length of hospital
stay, and a higher resource utili-
zation as measured by INPULS.
Patients with short-lasting delir-
ium had intermediate values as
compared to patients who were
never delirious. Even after con-
trolling for severity of illness,
time to first assessment, and
duration of mechanical ventila-
tion, delirium remained an inde-
pendent predictor of increased
hospital LOS (p = 0.043).

C D
DISCUSSION
Increasing incidence of ICD
reported in recent pediatric
studies may be understood
two-fold. First, without wide-
spread screening, the diagno-
sis of hypoactive delirium can
be missed easily even though
hypoactive symptoms may be
the most common presentation
of ICD (20). Pediatric delirium
tools such as the CAPD have
been specifically developed to
include symptoms of hypo-
active delirium. Second, the
majority of patients in the PICU
are younger than 2 years (5), a
group that was difficult to assess
in the past. With the develop-
Figure 2. Individual items of the Cornell Assessment of Pediatric Delirium (mean and sd) in children with no ment of suitable delirium tools
delirium (A) (white bars) and with severe delirium (B) (black bars). For children with mild delirium (gray bars), for all ages, significantly higher
two different diagrams are shown for asymptomatic (C) and symptomatic (D) periods, respectively. *Compared to
patients without delirium (A): p < 0.001. rates of ICD have been found
in younger children (7, 8).
Delirium in adults is predomi-
shows how the individual items of the CAPD reflect the DSM nantly found in the elderly, and it can be speculated that the
domains for the diagnosis of delirium. Items 2, 5, and 8 that developing as well as the ageing brain are especially sensitive
are affected in both group 1 and group 2 patients are mostly to noxious stimulation (21). In our collective, 57% of the chil-
related to cognition, orientation, affect, and psychomotor dren were younger than 2 years, 48% younger than 1 year, and
activity. Hence, the pattern found in group 1 can be distin- 34% younger than 6 months. In addition, most of our patients
guished from sedation and should be regarded as a mild or (75.3%) were ventilated during their PICU stay. The overall
short-lasting delirium. incidence of ICD in our study of 66% is the highest reported
so far for pediatric patients. It is in the range of the highest risk
Severity of Illness and Outcome Measures group in adults, that is, ventilated patients (9).
Clinical outcome of the study patients is given in Table 3. No sig- Studies in nongeriatric adult patients requiring intensive care
nificant difference in disease severity (PIM2) was found between show that the average duration of ICD is relatively short, typically

Pediatric Critical Care Medicine www.pccmjournal.org 131

Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
 Meyburg et al

Table 2. Items of the Cornell Assessment of Pediatric Delirium and Their Correspondence
to DSM Domains
Cornell Assessment of Pediatric Delirium Item Diagnostic and Statistical Manual Delirium Domains

Does the child make eye contact with the caregiver? Consciousness
Are the child’s actions purposeful? Cognition
Is the child aware of his/her surroundings? Consciousness
Orientation
Does the child communicate needs and wants? Consciousness
Psychomotor activity
Is the child restless? Orientation
Cognition
Affect/distress
Is the child inconsolable? Orientation
Cognition
Affect/distress
Is the child underactive—very little movement while awake? Orientation
Affect/distress
Does it take the child a long time to respond to interactions? Consciousness
Psychomotor activity

Table 3. Severity of Illness and Outcome Measures of Patients Without Delirium, With
Mild Delirium (< 24 Hr), and With Severe Delirium (> 24 hr)
No Delirium Mild Delirium Severe Delirium
Outcome Parameter (Group 0) (Group 1) (Group 2) p

n (%) 32 (34.4) 30 (32.3) 31 (33.3)

Pediatric Index of Mortality 2 1.21 ± 1.06 1.27 ± 0.84 1.41 ± 0.68 Not significant
Mechanical ventilation on the PICU (hr) 9.0 ± 18.7 29.7 ± 39.8 69.0 ± 74.5 < 0.001
Time to first analysis (d)
a
1.0 ± 1.0 2.0 ± 1.9 3.7 ± 3.2 < 0.001
Length of stay (d)
b
10.6 ± 7.3 13.5 ± 7.1 21.8 ± 19.7 0.043
Level of care (INPULS) 3.0 ± 0.3 2.9 ± 0.3 3.3 ± 0.4 < 0.001
Delirium screening commenced when Richmond Agitation and Sedation Score > –3.
a

Controlled for Pediatric Index of Mortality 2, duration of mechanical ventilation, and time to first assessment.
b

Values are expressed as mean ± sds.

ranging from 1 to 5 days (22, 23). If this is also the case in the
PICU, then screening for delirium should be initiated as early
as possible in the postoperative course. In fact, Smith et al (24)
noted that ICD had probably already resolved in a considerable
number of electable study patients due to the consenting process
that could have taken several days. In the present study, consent
was obtained prior to elective surgery, so assessment of delirium
could be started immediately once the RASS was greater than
–3. However, delirium may be missed not only in the beginning
but also in the later course due to the typical fluctuation of the
disorder. Unlike the pCAM-ICU/psCAM-ICU that yield brief, in
the moment assessments, the CAPD specifically takes the fluctu-
ating course of delirium into account. Nevertheless, Traube et al
(7) state that two of three of their few false negatives would have
been detected by performing the screening twice daily. This is
exactly what was done in the present study.
By frequent assessments at close intervals, we could dis-
tinguish two different patterns of ICD in our patients. Group
2 had full-blown delirium with permanently elevated CAPD
scores over several days. In a third of these children, the

132 www.pccmjournal.org February 2017 • Volume 18 • Number 2

Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited

symptoms did not resolve during the observation period of
5 days. Group 1 had a milder course with lower scores and a
maximum of two pathologic assessments in 5 days, typically
on the first day of the investigation. In adults, SSD is described,
where there is a partial presentation, with the presence of one
or more symptoms of delirium that do not meet the criteria for
a diagnosis of delirium (25). The milder course seen in half of
our delirious children (group 1) does not meet the definition
of SSD, as they had a full-blown, albeit short-lasting delirium.
What they share with adult patients classified as having SSD is
an intermediate outcome. Our group 1 patients took longer to
emerge from postoperative sedation as compared to children
who were never delirious, but again these periods were shorter
than in group 2. Also hospital LOS was prolonged in group 1
patients, but still shorter than in group 2.
Based on duration and intensity of symptoms, we would
therefore classify our patients as having no delirium (group 0),
mild delirium (group 1), or severe delirium (group 2). Since
studies in adult patients show that duration of postoperative
delirium is an independent risk factor for neurologic impair-
ment (23) and mortality (26), this classification may be impor-
tant for the design of follow-up studies in children.
A limitation of this study is that we do not discuss medica-
tions administered during the study period, as this is outside
the scope of this study. Future research assessing particular
medications used, and their effect on development of delir-
ium, is warranted. Another limitation of this study is that
delirium screening only continued for the first five postopera-
tive days. It is possible that with the passing of time, patients
in group 0 went on to develop delirium as iatrogenic risk fac-
tors accumulated. We were unable to account for this in the
current study design. Future longitudinal studies following
children throughout their PICU stay are necessary to further
characterize ICD.
CONCLUSIONS
In this study cohort, the majority of postoperative PICU
patients developed ICD, and early screening was necessary to
detect a substantial percentage of cases. In half of affected chil-
dren, delirium resolved within 24 hours and symptoms were
less severe. Development of delirium, and its duration, had a
measurable effect on hospital LOS. Our findings support the
description of delirium as a continuum of acute neurocognitive
disorder.
REFERENCES
1. Klein Klouwenberg PM, Zaal IJ, Spitoni C, et al: The attributable mor-
tality of delirium in critically ill patients: Prospective cohort study. BMJ
2014; 349:g6652
2. Pandharipande PP, Girard TD, Jackson JC, et al; BRAIN-ICU Study
Investigators: Long-term cognitive impairment after critical illness.
N Engl J Med 2013; 369:1306–1316
3. Prugh DG, Wagonfeld S, Metcalf D, et al: A clinical study of delirium
in children and adolescents. Psychosom Med 1980; 42:177–195
4. Jones SM, Fiser DH, Livingston RL: Behavioral changes in pediatric
intensive care units. Am J Dis Child 1992; 146:375–379
Pediatric Critical Care Medicine www.pccmjournal.org 133
Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
Feature Articles
5. Schieveld JN, Leroy PL, van Os J, et al: Pediatric delirium in critical
illness: Phenomenology, clinical correlates and treatment response
in 40 cases in the pediatric intensive care unit. Intensive Care Med
2007; 33:1033–1040
6. Janssen NJ, Tan EY, Staal M, et al: On the utility of diagnostic instru-
ments for pediatric delirium in critical illness: An evaluation of the
Pediatric Anesthesia Emergence Delirium Scale, the Delirium Rating
Scale 88, and the Delirium Rating Scale-Revised R-98. Intensive
Care Med 2011; 37:1331–1337
7. Traube C, Silver G, Kearney J, et al: Cornell Assessment of Pediatric
Delirium: A valid, rapid, observational tool for screening delirium in the
PICU. Crit Care Med 2014; 42:656–663
8. Smith HA, Gangopadhyay M, Goben CM, et al: The Preschool
Confusion Assessment Method for the ICU: Valid and reliable delir-
ium monitoring for critically ill infants and children. Crit Care Med
2016; 44:592–600
9. Ely EW, Inouye SK, Bernard GR, et al: Delirium in mechanically
ventilated patients: Validity and reliability of the confusion assess-
ment method for the intensive care unit (CAM-ICU). JAMA 2001;
286:2703–2710
10. Turkel SB, Tavaré CJ: Delirium in children and adolescents.
J Neuropsychiatry Clin Neurosci 2003; 15:431–435
11. Kelly P, Frosch E: Recognition of delirium on pediatric hospital ser-
vices. Psychosomatics 2012; 53:446–451
12. Smeets IA, Tan EY, Vossen HG, et al: Prolonged stay at the paediat-
ric intensive care unit associated with paediatric delirium. Eur Child
Adolesc Psychiatry 2010; 19:389–393
13. Silver G, Kearney J, Traube C, et al: Pediatric delirium: Evaluating the
gold standard. Palliat Support Care 2015; 13:513–516
14. Slooff VD, Spaans E, van Puijenbroek E, et al: Adverse events of halo-
peridol for the treatment of delirium in critically ill children. Intensive
Care Med 2014; 40:1602–1603
15. Dill ML, von Haken R, Traube C, et al: Diagnosis of delirium in
pediatric intensive care patients. Prospective study to establish
the German version of the CAPD. Monatsschr Kinderheilk 2016;
164:308–317
16. American Psychiatric Association: Diagnostic and Statistical Manual
of Mental Disorders. Fifth Edition. Arlington, VA, American Psychiatric
Association, 2013
17. Ely EW, Truman B, Shintani A, et al: Monitoring sedation sta-
tus over time in ICU patients: Reliability and validity of the
Richmond Agitation-Sedation Scale (RASS). JAMA 2003;
289:2983–2991
18. Slater A, Shann F, Pearson G; Paediatric Index of Mortality (PIM)
Study Group: PIM2: A revised version of the Paediatric Index of
Mortality. Intensive Care Med 2003; 29:278–285
19. Joa C, Faschingbauer C: [Calculating cost effectiveness in intensive
care nursing: INPULS (Intensive Nursing and Performance System)
clarifies expenditures]. Pflege Z 2003; 56:651–655
20. Peterson JF, Pun BT, Dittus RS, et al: Delirium and its motoric subtypes:
A study of 614 critically ill patients. J Am Geriatr Soc 2006; 54:479–484
21. Grill MF, Maganti RK: Neurotoxic effects associated with antibi-
otic use: Management considerations. Br J Clin Pharmacol 2011;
72:381–393
22. Bamalwa M, Mahmood SN, Praharaj SK: Delirium in cardiac ICU
patients. Ann Clin Psychiatry 2016; 28:51–55
23. Brummel NE, Jackson JC, Pandharipande PP, et al: Delirium in the
ICU and subsequent long-term disability among survivors of mechani-
cal ventilation. Crit Care Med 2014; 42:369–377
24. Smith HA, Boyd J, Fuchs DC, et al: Diagnosing delirium in critically ill
children: Validity and reliability of the Pediatric Confusion Assessment
Method for the Intensive Care Unit. Crit Care Med 2011; 39:150–157
25. Lipowski ZJ: Delirium: Acute Confusional States. New York, USA,
Oxford University Press, 1990
26. Shehabi Y, Riker RR, Bokesch PM, et al; SEDCOM (Safety and
Efficacy of Dexmedetomidine Compared With Midazolam) Study
Group: Delirium duration and mortality in lightly sedated, mechani-
cally ventilated intensive care patients. Crit Care Med 2010;
38:2311–2318