Society for Pediatric Anesthesia

Section Editor: Peter J. Davis

Cardiovascular Effects of Dexmedetomidine Sedation

in Children
Jackson Wong, MD,* Garry M. Steil, PhD,* Michelle Curtis, PNP,† Alexandra Papas, BS,‡
David Zurakowski, PhD,§ and Keira P. Mason, MD§

BACKGROUND: Dexmedetomidine (DEX) affects heart rate (HR), mean arterial blood pressure,
cardiac index (CI), stroke index (SI), and systemic vascular resistance index (SVRI) in adults. In this
study we sought to determine whether similar effects occur in children undergoing DEX sedation.
METHODS: Hemodynamic changes in children were followed during IV DEX sedation for
radiological procedures. One group of 8 patients (DEX-brief) received a bolus (2 mcg/kg bolus
over 10 minutes) and completed the procedure within 10 minutes. The second group of 9
patients (DEX-prolong) received the bolus plus additional DEX as needed to maintain sedation for
procedures lasting longer than 10 minutes (additional 1 mcg/kg/hr infusion with second bolus
if needed). CI, SI, and SVRI were measured using a continuous noninvasive cardiac output
monitor. Changes in hemodynamic variables at minutes 10, 20, and discharge (time at which
patient achieved Aldrete Score ⱖ9) were compared to baseline by repeated measures ANOVA
with effect sizes reported as mean [95% confidence interval].
RESULTS: Data were obtained during 8 DEX-brief and 9 DEX-prolong procedures. In DEX-brief, HR
and CI decreased (18.9 [2.3 to 35.5] bpm and 0.74 [0.15 to 1.33] L/min/m2; respectively) at T1.
There was no change in any other hemodynamic variables and all hemodynamic variables returned
to baseline at recovery. In DEX-prolong, both HR and CI remained decreased (24.0 [8.3 to 39.6] bpm,
1.51 [0.95 to 2.06] L/min/m2; respectively) at recovery. In addition, SI was decreased (8.01 [1.71 to
14.31] mL/m2) and SVRI was increased (776.0 [271.9 to 1280.4] dynes-sec/cm5/m2) at recovery in the
DEX-prolong group. There were no significant changes in mean arterial blood pressure in either group.
CONCLUSION: DEX decreases CI in children and has a cumulative effect. For patients
undergoing prolonged procedures HR and CI remained decreased at the time of discharge
together with a decrease in SI and an increase in SVRI. (Anesth Analg 2012;114:193–9)

D exmedetomidine (DEX) is an ␣-2 agonist that can

elicit marked fluctuations in heart rate (HR) and
mean arterial blood pressure (MAP) in adults and
children.1– 4 Approved by the Food and Drug Administra-
tion (FDA) in 2008 for adult sedation outside of the
operating room, it is still not approved for pediatric use.4
As a sedative, it has a respiratory-sparing effect, although it
can elicit changes in HR and MAP that sometimes necessi-
tate medical intervention.5–9 Studies in healthy adults have
shown incremental increases in DEX to elicit incremental
decreases in HR and MAP, coincident with decreases in
From the *Department of Medicine, Medicine Critical Care Program,
Children’s Hospital Boston and Harvard Medical School, Boston, MA,
†Children’s Hospital Boston, Boston, MA, ‡Children Hospital Boston and
Harvard Medical School, Boston, MA, and §Department of Anesthesia,
Children Hospital Boston and Harvard Medical School, Boston, MA.
Accepted for publication August 3, 2011.
Jackson Wong, MD has a Clinical Trial Agreement with Cardiotronic Inc, LA
Jolla CA. Cardiotronic Inc provided funding for this investigator initiative
study. The research proposal, design, conduct of the study, data analysis,
and manuscript were performed solely by the investigators at Children’s
Hospital Boston.
Conflict of Interest: See Disclosures at the end of the article.
Reprints will not be available from the authors.
Address correspondence to Jackson Wong, MD, Children’s Hospital Boston
and Harvard Medical School, Department of Medicine, Medicine Critical
Care Program; 300 Longwood Avenue 11 South, Boston, MA 02115. Address
e-mail to Jackson.Wong@childrens.harvard.edu
Copyright © 2012 International Anesthesia Research Society
DOI: 10.1213/ANE.0b013e3182326d5a
cardiac index (CI) and stroke index (SI), and an increase in
systemic vascular resistance index (SVRI).10 –17 Investigations
in noncritically ill children have also shown HR and MAP to
decrease during DEX sedation1,18,19 and its effect on other
cardiovascular variables has been described in critically ill
children.20 –22 However, the effects on CI, SI, and SVRI in
children without cardiac disease undergoing sedation for
radiological imaging studies have not been described.
In this study, a continuous noninvasive cardiac output
(CO) monitor was used to assess the changes in hemody-
namic variables in children undergoing sedation not re-
quiring invasive central catheters or mechanical ventilation.
The CO monitor (ICON monitor; Cardiotronic Inc., La Jolla,
CA) is FDA-approved for infants, children, and adults and
has been studied in children with congenital heart disease
anesthetized during cardiac catheterization.23,24 In a study of
32 pediatric patients,23 aged 11 days to 17.8 years, the monitor
was shown to correlate well with the Fick method to obtain
CO. A study of 50 children with heart disease,24 aged 0.5 to
16.5 years, showed the monitor to correlate well with CO
measurements obtained by thermodilution. The primary aim of
the present study was to use the monitor to determine whether
DEX acutely affects CI, SI, and SVRI in children undergoing
sedation and to determine how long the effects persist.
METHODS
This prospective observational study was approved by the
Committee on Clinical Investigation for human research
(IRB) at Children’s Hospital Boston. All subjects and/or

January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 193

Cardiovascular Effects of Dexmedetomidine
Table 1. Exclusion Criteria
Current treatment with oral clonidine
Patients with an ASA score ⱖ3
Pacemakers and vagus nerve stimulator
Mitral or aortic valve dysfunction
Dextrocardia
Second- or third-degree heart block
Current diagnosis of cardiac, pulmonary, hepatic or renal failure
Current diagnosis of pulmonary masses/tumor/pleural effusions/
pneumonia/edema
Pericardial effusion
Concomitant use of hypertension medications including angiotensin
converting enzyme inhibitors, beta receptor and calcium channel
blockers.
Large implanted metallic devices (including orthodontic braces, spine
rods, plates and screws)
legal guardians signed informed consent to participate.
During the study, the medical staff, nurses, family, subject,
and research team were blinded to real-time results. The
study is registered with the Clinical Trial Registry
(NCT01001533 10/23/2009).
Inclusion Criteria
All pediatric patients (⬍18 years of age) and ⱕASA 2
scheduled to receive DEX sedation per institutionally ap-
proved dosing protocol18,19 for elective computerized to-
mography and nuclear medicine scan procedures were
eligible to participate.
Exclusion Criteria
Medical conditions that contraindicate DEX per institutionally
approved dosing protocol18 and patient/CO-monitor–related
exclusion criteria are shown in Table 1.
DEX Sedation Protocol
All patients were Nil Per Os for a minimum of 2 h of clear
liquids and 6 h of solids. No premedications for anxiolysis
or sedation were administrated before receiving DEX. All
children received IV DEX per institutionally approved
protocol: 2 mcg/kg bolus over 10 minutes followed by a
continuous infusion of 1 mcg/kg/hour until the procedure
was completed. Per protocol, the bolus (2 mcg/kg over 10
minutes) was repeated in the event of failure to achieve
adequate sedation or to maintain a Ramsay Sedation Score
(RSS) of 4.25 As soon as the patient achieved a minimum
RSS of 4 the procedure was initiated.18 If this occurred
before completion of the bolus, the scan was initiated as the
bolus was completed. All boluses were completed. Patients
were divided into 2 groups depending on the duration of
the procedure. Children undergoing procedures lasting less
than or equal to 10 min were grouped as DEX-brief.
Children undergoing procedures lasting ⬎10 minutes were
grouped as DEX-prolong.
All patients were continuously monitored for oxygen
saturation (pulse oximetry) and HR. Noninvasive MAP
was recorded at 5-minute intervals. In the event of a
decrease in MAP ⬎20% below an age-adjusted awake
normal value,26 an IV saline bolus of 10 mL/kg was
administered per institutional protocol.18,26 After the
completion of the procedure, all patients were transferred
to the radiology recovery room where, per protocol, a 20
194 www.anesthesia-analgesia.org
mL/kg normal saline solution (NSS) bolus was adminis-
tered on arrival. A subsequent decrease in arterial blood
pressure to 20% below age-adjusted awake normal values
triggered an additional 10 mL/kg NSS bolus, per protocol.
Physiological monitoring was continued and, along with
RSS, documented at 5-minute intervals until discharge.
Discharge criteria18,26 were defined as patient having
achieved and maintained an Aldrete Score27 ⱖ9 for a
minimum of 30 min.
Hemodynamic Monitoring
CO was monitored using a portable noninvasive continu-
ous CO monitor approved by the FDA for infants, children,
and adults (ICON monitor; Cardiotronic Inc., La Jolla, CA).
The monitor uses electrical cardiometry, a subtype of
bio-impedance,23,24,28,29 and requires standard electrocar-
diogram electrodes (Vermed Inc., Bellows Falls, VT) be
placed on the left side of the body at (A) the level of the
tragus, (B) 5 cm below sensor A, (C) left mid-axillary line at
the level of the xiphoid process, and (D) 10 cm below sensor
C. Subject weight (kg) and height (cm) are entered into the
monitor to calculate body surface area (BSA), CI (CI ⫽
CO/BSA) and SI (SI ⫽ stroke volume/BSA) over 10 cycle
intervals.
Data Acquisition
Data collection was initiated 5 minutes before commencing
DEX and continued until 5 minutes after discharge criteria
were met. The CO monitor signal strength was recorded
during the procedure on a clinical report form. After each
procedure, CO monitor data (time, HR, CI, and SI) were
downloaded to a computer file and merged with routine
vital signs (systemic arterial blood pressure [SAP] and
diastolic arterial blood pressure [DAP]), and medical inter-
vention records (DEX, NSS and pharmacologic interven-
tions for HR and MAP). Patient demographics (age, gender,
weight and height), type of scan (imaging computed to-
mography or nuclear medicine scan), RSS, and Aldrete
Score were all obtained from medical, sedation, and quality
assurance records. Data were compiled with Microsoft
Access version 3.0 (Microsoft Inc, Redmond, WA).
Data Analysis and Statistics
HR, CI, and SI were compared at 4 time points: 0 minutes,
defined as baseline (BL); 10 minutes (T1); 20 minutes (T2); and
recovery (R). For continuous measurements of CI, SI, HR,
comparisons were made using a 5-minute average for each
time point: 5-minute interval immediately before starting
DEX for BL; 5 to 10 minutes after the start of DEX for T1; 15 to
20 minutes for T2; and the 5-minute interval immediately after
discharge criteria was met for R.
MAP was calculated as (2 ⫻ DAP ⫹SAP)/3 using the
DAP and SAP measures obtained at the 5-minute sched-
uled sample interval. SVRI (dynes-sec/cm5/m2) was calcu-
lated as [MAP-CVP)/CI] ⫻ 80 where the central venous
pressure (CVP) taken to be 4 mm Hg.30 The sensitivity of
the SVRI calculation was then assessed by reanalyzing the
data assuming CVP values of below (0 and 2 mm Hg) and
above (6 and 8 mm Hg) this value. In instances in which
MAP was not obtained at T1 or T2, the missing MAP value
ANESTHESIA & ANALGESIA
was interpolated using 2 adjacent 5-minute values, if avail-
able, or replaced with an adjacent 5-minute value if adjacent
values on both sides were not available. Missing values (⬃7%)
were noted and a subsequent error analysis was conducted to
assess the impact of imputing the missing data. For the error
analysis, the mean absolute relative difference (MARD) be-
tween imputed values and measured was calculated for all
time points in which the MAP was available.
Changes from BL in the continuous measurement ob-
tained from the CO monitor (HR, CI, and SI) were assessed
using repeated-measures ANOVA using lag 1 auto-
regressive covariance structure with time (BL, T1, T2, and
R) and length of procedure (DEX-brief and DEX-prolong)
as main effects. Post hoc analysis was performed using
Fisher’s least significant difference, which was chosen to
minimize type II statistical error (increase the power to
detect an adverse effect of DEX on any of the measured
hemodynamic variables) rather than a more conservative
procedure protecting for type I error. An internal Signal
Quality Index (SQI) was used to assess signal quality, with
SQI ⱖ70 (7 of 10 cycles having the same value) taken as an
indicator of reliability. The monitor also reports signal
strength with 3 of 5 bars indicating a good signal and 4 of
5 bars indicating an excellent signal (personal communica-
tion, Dr. Markus Osypka, Cardiotronic Inc., 2009). These
values were also recorded on clinical report forms. Subjects
with insufficient CO signal strength or disconnected leads,
in any of the 4 time intervals, were excluded from analysis.
Changes in the noncontinuous measurements (MAP and
SVRI) were likewise tested with the 2-way repeated-
measures ANOVA.
Patient demographics (age, weight, and height) are
reported as median and range, with comparison between
DEX-brief and DEX-prolong made with Mann-Whitney
tests. Gender differences were compared using ␹2 tests.
Differences in the time to meet discharge criteria after DEX
Table 2. Demographics
DEX-brief DEX-prolong
Total subjects 8 9
Gender (male/female) 4/4 7/2
Age (year) 2.8 (range 0.3–6.1) 2.4 (range 1.2–4.3)
Weight (kg) 13.5 (range 6–20) 13.0 (range 11–17)
Height (cm) 92 (range 67–107) 86 (range 78–106)
Children undergoing procedures less than 10 min in duration (dexmedetomi-
dine 关DEX兴-brief) received 2 mcg/kg DEX over 10 minutes. Children with
procedures longer than 10 min (DEX-prolong) received the initial 2 mcg/kg of
DEX plus additional DEX as need to obtain and maintain a Ramsay Sedation
Score of 4. No significant differences was found between the 2 treatment
groups (P⬎0.05 all).
Table 3. Time Intervals and Additional Normal Saline Boluses
Mean time (hours) between last oral clear liquid intake to first dose of DEX (NPO)
Mean time of DEX administration (minutes)
Mean time to discharge criteria after discontinuation of dexmedetomidine (minutes)
Number of subjects requiring additional normal saline bolus (10 cc/kg) for hypotension
Children undergoing brief procedures (dexmedetomidine 关DEX兴-brief) received a bolus of DEX given over 10 minutes. Children with procedures longer than 10
minutes (DEX-prolong) received the bolus plus additional dexmedetomidine as needed. Difference in time of administration was not tested (longer in the prolonged
procedure). *Difference in the time to achieve discharge criteria after discontinuing DEX was not significant (7 关-25 to 39兴 minutes). †Differences in the percent
of subjects receiving additional normal saline bolus was not significant (19.4 关–29% to 63%兴).
January 2012 • Volume 114 • Number 1
sedation were compared using Student’s t-tests with un-
equal variances. Data in figures are reported as mean ⫾
SEM. Differences from BL, or between groups, are reported
as mean [95% confidence interval]. A post hoc analysis of
the power to detected changes in CI was performed.31
Statistical analysis was performed using PASW Statistics
Version 18 for Windows (SPSS Inc., Chicago, IL; mixed-
model repeated-measures ANOVA), GraphPad Prism
Version 5.04 (GraphPad Software, Inc, La Jolla CA; Mann-
Whitney tests, ␹2 tests and t-test with unequal variances),
and StatXact Version 9 (Cytel Software Corporation, Cam-
bridge, MA; 95% confidence interval for difference in
proportions).
RESULTS
Thirty-three subjects were enrolled in the study. Five
subjects did not require sedation, 1 withdrew and 1 was
excluded due to receiving pentobarbital as an adjuvant
sedative. In 9 additional subjects, CO measurements could
not be obtained in all time intervals due to problems with
the device leads (7) or cables (2), leaving 17 subjects with
data at all time points: 8 undergoing DEX-brief procedures
and 9 DEX-prolong procedures. All of these subjects had
either good or excellent signal strength at all time intervals,
with 93% of measurements having an SQI ⱖ70 (87% at BL,
88% at T1, 98% at T2, and 95% in R). MAP could not be
obtained at T1 or T2 in 2 DEX-brief and 3 DEX-prolong
subjects, with adjacent or interpolated 5-minute values
used to impute 7 of 68 scheduled recordings (7.4%). The
mean error introduced by this method for imputed values
was estimated not to be different from 0 (P ⫽ 0.86 and 0.92
for DEX-brief and DEX-prolong groups; estimate obtained
by taking the mean difference between imputed and mea-
sured values at time points where the MAP measure was
available). The error introduced by imputing missing MAP
values was estimated to be 5% (MARD) in the DEX-brief
group and 11.5% in DEX-prolong group.
Gender, age, weight, and height were similar in subjects
undergoing the DEX-brief and DEX-prolong procedures
(Table 2). Of the 9 patients in the DEX-prolong procedure
group, 3 had RSS ⬍4 at T1 and received a second 2 mcg/kg
of DEX. The mean time between the last oral clear liquid
intake (Nil Per Os) and the time of the first DEX bolus, the
duration of DEX administration, the mean time to achieve
discharge criteria, and the number of NSS boluses in
response to hypotension are reported in Table 3. No
patients received pharmacologic therapy for cardiovascular
fluctuations in HR or MAP.
Significant differences were observed in CI by group
(P ⫽ 0.043) and time (P⬍0.001), with the effect by time
DEX-brief DEX-prolong
4.2 (range 2.1–12.1) 7.0 (range 2.5–14.5)
10 23 (range 20–110)
65 (range 18–119) 49 (range 15–102)*
2/8 (25%) 4/9 (44.4%)†
www.anesthesia-analgesia.org 195
Cardiovascular Effects of Dexmedetomidine

Table 4. Two-Way Repeated Measures ANOVA

Group effect Time effect Group ⴛ Time interaction
Variable F-test P value F-test P value F-test P value
CI 4.89 0.043 17.27 ⬍0.001 2.98 0.042
HR 0.09 0.774 20.37 ⬍0.001 1.03 0.387
MAP 6.34 0.018 1.09 0.364 0.33 0.806
SI 5.74 0.030 4.09 0.012 1.12 0.351
SVRI 2.86 0.111 3.66 0.020 1.24 0.309
Group denotes children undergoing brief (dexmedetomidine 关DEX兴-brief, ⱕ 10 minutes) vs prolonged (DEX-prolonged, ⬎10 min) DEX procedures. Each variable (CI
indicates cardiac index; HR, heart rate; MAP, mean arterial blood pressure; SI, stroke index; SVRI, systemic vascular resistance index) was assessed at 4 different
time points (baseline defined as 0 min; T1 ⫽ 10 min, T2 ⫽ 20 min, and recovery, defined as the time at which discharge criteria was achieved).

108.7 at BL to 89.8 bpm at T1 (18.9 [2.3 to 35.5]) and

Figure 1. Time course of the effects of dexmedetomidine (DEX) on
heart rate (HR; mean ⫾ SEM), cardiac index (CI), and stroke index
(SI) in children undergoing brief (DEX-brief; Blue; ⱕ10 min) and
prolonged (DEX-prolong; Red, ⬎10 min) procedures. Time course is
in minutes with baseline (BL) defined as minute 0 (start of DEX
bolus; 2 mcg/kg over 10 min); T1 defined as minute 10; T2 defined
as minute 20 (end of second bolus or 1 mcg/kg/h infusion as
needed), and R defined at the time at which discharge criteria was
met (varies by patient). Differences from BL are indicted by †
(DEX-brief) and ‡ (DEX-prolong); differences between groups by
*(P ⬍ 0.05, 2-way repeated measures ANOVA with least significant
difference post hoc analysis).
different in the 2 groups (interaction, P ⫽ 0.042; Table 4). In
all remaining variables (HR, MAP, SI, and SVRI), there was
either a significant effect by group or a significant interac-
tion (Table 4). During the DEX-brief, HR decreased from
returned to BL at R (9.9 [– 6.7 to 26.5] bpm). This was similar
to the decrease observed with the DEX-prolong procedure
where HR decreased from 117.9 at BL to 95.7 bpm at T1
(22.2 [6.6 to 37.9]), and remained decreased to BL at R (24.0
[8.3 to 39.6] bpm) with the difference at R not different
compared to DEX-brief (4.9 [-21.0 to 11.2] bpm; Fig. 1A).
Similarly, CI decreased from BL to T1 in both the DEX-brief
(3.96 to 3.21 or 0.74 [0.15 to 1.34] L/minutes/m2) and
DEX-prolong (3.98 to 2.96 or 1.02 [0.46 to 1.58]
L/minutes/m2) groups with the difference at T1 not differ-
ent (0.26 [– 0.32 to 0.83] L/minutes/m2). CI returned to BL
in the DEX-brief group at R (0.44 [-0.15 to 1.03]
L/minutes/m2); however, it did not return to BL in the
DEX-prolong group (1.51 [0.95 to 2.06] L/minutes/m2),
with the decrease at R more pronounced in the prolonged
procedure (1.04 [0.47 to 1.62] L/minutes/m2; Fig. 1B). SI
was not decreased at R relative to BL in the DEX-brief
group (37.51 to 36.32 or 1.19 [–5.50 to 7.87] mL/m2) but was
significantly decreased in the DEX-prolong group (34.60 to
26.58 or 8.01 [1.71 to 14.31] mL/m2), with the value at R
significantly lower than in the DEX-brief procedure (9.74
[3.25 to 16.24] mL/m2) (Fig. 1C).
MAP did not change from BL in either the DEX-brief or
the DEX-prolong groups (Fig. 2A). SVRI, assuming a CVP
value of 4 mm Hg, did not change from BL in the DEX-brief
procedure group but increased from 1424.0 to 2200.2
(776.2 [271.9 to 1280.4] dynes-sec/cm5/m2; Fig. 2B) in the
DEX-prolong group, with the value at R being signifi-
cantly higher compared to DEX-brief (663.0 [143.3 to
1182.8] dynes-sec/cm5/m2). Similar results were ob-
tained for CVP values of 0, 2, 6, and 8 mm Hg, because
these changes in CVP do not affect the calculated value
of SVRI by ⬎6%.
Nine subjects were removed from the repeated-
measures analysis due to inability to obtain data at all time
points. The inability to obtain data was largely due to the
children removing the CO leads (7 patients) as they were
waking up from the sedation. To assess the impact this may
have had on our power to detect differences in CI, we
performed a post hoc analysis of statistical power. Based on
data obtained in the 17 subjects who completed the proce-
dure, we estimate that our analysis had 82% power for
detecting differences in CI (noncentral F ⫽ 11.04, partial
␩2 ⫽ 0.272) and 78% power for detecting changes in HR
(noncentral F ⫽ 8.54, partial ␩2 ⫽ 0.363). Power for estimat-
ing changes from BL was ⬎99% for both CI and HR

196 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA


Figure 2. Effects of dexmedetomidine (DEX) on mean arterial blood
pressure (MAP; mean ⫾ SEM) and systemic vascular resistance
index (SVRI) in the DEX-brief (Blue) and DEX-prolong (Red) groups.
Baseline (BL) defined as minute 0 (start of dexmedetomidine bolus;
2 mcg/kg over 10 min); T1 defined as minute 10; T2 defined as
minute 20 (end of second bolus or 1 mcg/kg/h infusion as needed),
and R defined at the time at which discharge criteria was met (varies
by patient). Differences from BL are indicted by † (DEX-brief) and ‡
(DEX-prolong); differences between groups by *(P ⬍ 0.05, 2-way
repeated measures ANOVA with least significant difference post hoc
analysis).
(Eta-squared ⫽ 0.57, noncentral F ⫽ 39.9; and ␩2 ⫽ 0.67,
noncentral F ⫽ 73.3, respectively).
DISCUSSION
The present study examined the hemodynamic effects of
DEX sedation in healthy children undergoing radiological
imaging studies. Consistent with our previous study19 we
found a significant decrease in HR within 5 to 10 minutes of
initiating a DEX bolus. In this study, we show that HR and
CI decreased after a single bolus and recovered to BL
within 60 minutes after a brief exposure. No changes in SI
or SVRI were observed after a brief exposure, but pro-
longed exposure leads to decreases in HR, CI, and SI that
did not recover back to BL. This was accompanied by a
corresponding increase in SVRI in the group with pro-
longed exposure of DEX. These effects can persist for up to
1 hour after stopping DEX infusions. No significant
changes in MAP were observed after either brief or pro-
longed exposure.
Our results are consistent with adult studies demon-
strating that HR, MAP, and CI can remain decreased for up
to 4 hours after stopping DEX.10 The increase in SVRI and
normal MAP observed in our subjects is similar to the
hemodynamic pattern in adults.10 The pharmacokinetics of
DEX in adults and children appear to be similar at low
doses,32–35 albeit higher clearance rates have been observed
in neonates32–35 and the kinetics at the higher dose used in
January 2012 • Volume 114 • Number 1
this study have not been studied. HR decreased during
sedation but this was not always indicative of the change in
CI (Fig. 1). In the DEX-prolong procedure group, HR was
decreased 20% from BL when discharge criteria was
achieved, accompanied by a 38% decrease in CI and a 23%
decrease in SI. These hemodynamic changes are consistent
with those observed in adult sedation studies using invasive
thermodilution and pulse contour CO analysis.10,12,13,16,17
Studies in adults using bio-impedance to assess CI have
also reported similar decreases, with reported values for CI
decreasing from 17%–38%.11,14,15 The decrease in CI found
in our study was dose-dependent, with more pronounced
effects being observed during prolonged procedures in
which more DEX was used to maintain sedation. We found
that CI acutely decreases after a single 2 mcg/kg DEX bolus
and that in 3 patients who required a second bolus to
achieve an RSS of 4, generally accepted as an adequate level
for radiological imaging, the effect was more pronounced.
Although this was not tested for statistical significance due
to the low number of subjects (n ⫽ 3), it supports our
conclusion that the effect is dose-dependent. Lower doses
of DEX have been argued to not affect left ventricle function
in postcardiac surgery children,20 although these patients
may have received vasoactive medications. An unantici-
pated finding in the present study was that the decrease in
SI in the DEX-prolong subjects persisted up to the time
discharge criteria was met, in some cases up to 1 hour after
discontinuation of the DEX (Fig. 1C). This effect has not
been observed in adults, where SI typically returns to BL
within 50 minutes of stopping DEX.10 Further studies to
determine the etiology of the decrease in SI are required
because SI is dependent on multiple variables.36
The use of a noninvasive CO monitor is a limitation
in the present study. Use of invasive measures such as Fick
or thermodilution may allow more precise data to be
obtained, albeit not continuously. However, because our
patients were free of chronic respiratory and cardiac dis-
ease, and were undergoing sedation outside of the inten-
sive care unit, the invasive methods may not have been
appropriate. The CO monitor we used has been shown to
correlate with both Fick and thermodilution CO method in
children during general anesthesia.23,24 We noted that the
electrocardiogram leads, particularly those applied at the
neck, were not tolerated by all children, with 7 patients
having removed the leads before complete acquisition of
data. A post hoc power analysis indicated that the loss of
these subjects did not substantially impact our power,
which was about 80% for detecting differences in CI
between groups undergoing brief versus prolonged proce-
dures, and ⬎98% for detecting differences from BL within
each group.
The inability to obtain MAP readings in 7% of our
scheduled blood pressure readings may also have limited
our ability to detect differences in this variable. To avoid
excluding these subjects, we interpolated the missing data
using adjacent 5-minute measurements, or substituted 1
adjacent 5-minute value if 5-minute values were not avail-
able both before and after the scheduled measurement. A
retrospective error analysis of the error that could be
introduced using this method suggested that the imputed
value is unbiased (expected error not different from zero)
www.anesthesia-analgesia.org 197
Cardiovascular Effects of Dexmedetomidine
but that the expected error in any one measure would be
between 6% and 11% (MARD). This may have reduced our
power to detect any subtle differences in MAP. We have
previously evaluated changes in MAP in a larger cohort of
subjects (n ⫽ 198), using the identical protocol to that
studied here, and shown it to decrease by 10% to 15%.19
A further limitation to the present study is the absence
of a direct measure of the CVP used to calculate SVRI. We
assumed a value of 4 mm Hg. Normal values for CVP
reported in the literature for children range from 0 to 5
mm Hg,30 although values after cardiac surgery have been
reported to be as high as 10 mm Hg.20,37,38 CVP has not
been shown to change in response to DEX.20,37 To assess the
impact of assuming a value of 4 mm Hg, we reanalyzed all
data assuming CVP values between 0, 2, 4, 6, and 8 mm Hg
and found the differences in the calculated SVRI to be no
⬎6% even if CVP was increased by 4 to 8 mm Hg (100%)
secondary to fluid administration.
In summary, this is the first study to report the effect of
DEX sedation on CI, SI, and SVRI in children without
cardiac disease during radiological procedures. We found
that DEX decreases CI and SI with a concomitant increase
in SVRI. We found the effects to be dose-dependent and to
persist up to 1 hour after discontinuing the sedative.
DISCLOSURES
Name: Jackson Wong, MD.
Conflicts: Jackson Wong received research funding from Car-
diotronic Inc, La Jolla, CA for this investigator-initiated study.
Contribution: Jackson Wong has seen the original study data,
reviewed the analysis of the data, approved the final manu-
script, and is the author responsible for archiving the study
files.
Name: Garry M. Steil, PhD.
Contribution: This author helped design the study, conduct
the study, analyze the data, and write the manuscript.
Conflicts: None.
Attestation: Garry M. Steil has seen the original study data,
reviewed the analysis of the data, and approved the final
manuscript.
Name: Michelle Curtis, PNP.
Contribution: This author helped conduct the study and write
the manuscript.
Conflicts: None.
Attestation: Michelle Curtis has seen the original study data,
reviewed the analysis of the data, and approved the final
manuscript.
Name: Alexandra Papas, BS.
Contribution: This author helped conduct the study, analyze
the data, and write the manuscript.
Conflicts: None.
Attestation: Alexandra Papas has seen the original study data,
reviewed the analysis of the data, and approved the final
manuscript.
Name: David Zurakowski, PhD.
Contribution: Statistician.
Conflicts: None.
Attestation: David Zurakowski has seen the original study
data, reviewed the analysis of the data, and approved the final
manuscript.
Name: Keira P. Mason, MD.
Contribution: This author helped design the study, conduct
the study, analyze the data, and write the manuscript.
198 www.anesthesia-analgesia.org
Conflicts: None.
Attestation: Keira P. Mason has seen the original study data,
reviewed the analysis of the data, and approved the final
manuscript.
This manuscript was handled by: Peter J. Davis, MD.
ACKNOWLEDGMENTS
The authors would like to express their gratitude to Amanda
Buckley and Melissa Whalen for their editorial and adminis-
trative assistance in preparation of this manuscript.
REFERENCES
1. Tobias JD. Dexmedetomidine: applications in pediatric critical
care and pediatric anesthesiology. Pediatr Crit Care Med
2007;8:115–31
2. Phan H, Nahata MC. Clinical uses of dexmedetomidine in
pediatric patients. Paediatr Drugs 2008;10:49 – 69
3. Gerlach AT, Murphy CV, Dasta JF. An updated focused review
of dexmedetomidine in adults. Ann Pharmacother 2009;43:
2064 –74
4. Su F, Hammer. Dexmedetomidine: pediatric pharmacology,
clinical uses and safety. Expert Opin Drug Saf 2011;10:55– 66
5. Sichrovsky TC, Mittal S, Steinberg JS. Dexmedetomidine seda-
tion leading to refractory cardiogenic shock. Anesth Analg
2008;106:1784 – 6
6. Zhang X, Schmidt U, Wain JC, Bigatello L. Bradycardia
leading to asystole during dexmedetomidine infusion in an
18 year-old double-lung transplant recipient. J Clin Anesth
2010;22:45–9
7. Gerlach AT, Murphy CV. Dexmedetomidine-associated brady-
cardia progressing to pulseless electrical activity: case report
and review of the literature. Pharmacotherapy 2009;29:1492
8. Nagasaka Y, Machino A, Fujikake K, Kawamoto E, Wakamatsu
M. [Cardiac arrest induced by dexmedetomidine]. Masui
2009;58:987–9
9. Kato J, Ogawa Y, Kojima W, Aoki K, Ogawa S, Iwasaki K.
Cardiovascular reflex responses to temporal reduction in arte-
rial pressure during dexmedetomidine infusion: a double-
blind, randomized, and placebo-controlled study. Br J Anaesth
2009;103:561–5
10. Ebert TJ, Hall JE, Barney JA, Uhrich TD, Colinco MD. The
effects of increasing plasma concentrations of dexmedetomi-
dine in humans. Anesthesiology 2000;93:382–94
11. Bloor BC, Ward DS, Belleville JP, Maze M. Effects of intrave-
nous dexmedetomidine in humans. II. Hemodynamic changes.
Anesthesiology 1992;77:1134 – 42
12. Snapir A, Posti J, Kentala E, Koskenvuo J, Sundell J, Tuunanen
H, Hakala K, Scheinin H, Knuuti J, Scheinin M. Effects of low
and high plasma concentrations of dexmedetomidine on myo-
cardial perfusion and cardiac function in healthy male subjects.
Anesthesiology 2006;105:902–10; quiz 1069 –70
13. Venn M, Newman J, Grounds M. A phase II study to evaluate
the efficacy of dexmedetomidine for sedation in the medical
intensive care unit. Intensive Care Med 2003;29:201–7
14. Prielipp RC, Wall MH, Tobin JR, Groban L, Cannon MA, Fahey
FH, Gage HD, Stump DA, James RL, Bennett J, Butterworth J.
Dexmedetomidine-induced sedation in volunteers decreases
regional and global cerebral blood flow. Anesth Analg
2002;95:1052–9
15. Basar H, Akpinar S, Doganci N, Buyukkocak U, Kaymak C,
Sert O, Apan A. The effects of preanesthetic, single-dose
dexmedetomidine on induction, hemodynamic, and cardiovas-
cular parameters. J Clin Anesth 2008;20:431– 6
16. Penttila J, Snapir A, Kentala E, Koskenvuo J, Posti J, Scheinin
M, Scheinin H, Kuusela T. Estimation of cardiac output in a
pharmacological trial using a simple method based on arterial
blood pressure signal waveform: a comparison with pulmo-
nary thermodilution and echocardiographic methods. Eur
J Clin Pharmacol 2006;62:401–7
17. Ozkose Z, Demir FS, Pampal K, Yardim S. Hemodynamic and
anesthetic advantages of dexmedetomidine, an alpha
2-agonist, for surgery in prone position. Tohoku J Exp Med
2006;210:153– 60
ANESTHESIA & ANALGESIA
18. Mason KP, Zgleszewski SE, Dearden JL, Dumont RS, Pirich
MA, Stark CD, D’Angelo P, Macpherson S, Fontaine PJ,
Connor L, Zurakowski D. Dexmedetomidine for pediatric
sedation for computed tomography imaging studies. Anesth
Analg 2006;103:57– 62
19. Mason KP, Zgleszewski SE, Prescilla R, Fontaine PJ, Zura-
kowski D. Hemodynamic effects of dexmedetomidine sedation
for CT imaging studies. Paediatr Anaesth 2008;18:393– 402
20. Lazol JP, Lichtenstein SE, Jooste EH, Shiderly D, Kudchadker
NA, Tatum GH, Orr RA, Wearden PD, Morell VO, Munoz RA,
Chrysostomou C. Effect of dexmedetomidine on pulmonary
artery pressure after congenital cardiac surgery: A pilot study.
Pediatr Crit Care Med 2010;11:589 –92
21. Chrysostomou C, Sanchez De Toledo J, Avolio T, Motoa MV,
Berry D, Morell VO, Orr R, Munoz R. Dexmedetomidine use in
a pediatric cardiac intensive care unit: can we use it in infants
after cardiac surgery? Pediatr Crit Care Med 2009;10:654 – 60
22. Tokuhira N, Atagi K, Shimaoka H, Ujiro A, Otsuka Y, Ramsay
M. Dexmedetomidine sedation for pediatric post-Fontan pro-
cedure patients. Pediatr Crit Care Med 2009;10:207–12
23. Norozi K, Beck C, Osthaus WA, Wille I, Wessel A, Bertram H.
Electrical velocimetry for measuring cardiac output in children
with congenital heart disease. Br J Anaesth 2008;100:88 –94
24. Tomaske M, Knirsch W, Kretschmar O, Woitzek K, Balmer C,
Schmitz A, Bauersfeld U, Weiss M. Cardiac output measure-
ment in children: comparison of Aesculon cardiac output
monitor and thermodilution. Br J Anaesth 2008;100:517–20
25. Ramsay MA, Savege TM, Simpson BR, Goodwin R. Controlled
sedation with alphaxalone-alphadolone. BMJ 1974;2:656 –9
26. The Acutely Ill Child. In: Behrman RE, Kliegman RM, Jenson
HB, eds. Nelson Textbook of Pediatrics. 17th ed. Philadelphia:
Saunders, 2004:280
27. Aldrete JA, Kroulik D. A postanesthetic recovery score. Anesth
Analg 1970;49:924 –34
28. Zoremba N, Bickenbach J, Krauss B, Rossaint R, Kuhlen R,
Schalte G. Comparison of electrical velocimetry and thermodi-
lution techniques for the measurement of cardiac output. Acta
Anaesthesiol Scand 2007;51:1314 –9
January 2012 • Volume 114 • Number 1
29. Osypka MJ, Bernstein DP. Electrophysiologic principles and
theory of stroke volume determination by thoracic electrical
bioimpedance. AACN Clin Issues 1999;10:385–99
30. Normal Values and PRISM and TISS Scores. In: Toro-Figueroa
L, Levin D, Morriss F, eds. Essentials of Pediatric Intensive
Care Manual. St. Louis, MO: Quality Medical Publishing, Inc.,
1992:447
31. Murphy K, Myors B, Wollach A. Statistical power analysis: a
simple and general model for traditional and modern hypoth-
esis tests. New York: Taylor & Francis Group, 2009
32. Potts AL, Anderson BJ, Holford NH, Vu TC, Warman GR.
Dexmedetomidine hemodynamics in children after cardiac
surgery. Paediatr Anaesth 2010;20:425–33
33. Potts AL, Anderson BJ, Warman GR, Lerman J, Diaz SM, Vilo
S. Dexmedetomidine pharmacokinetics in pediatric intensive
care–a pooled analysis. Paediatr Anaesth 2009;19:1119 –29
34. Vilo S, Rautiainen P, Kaisti K, Aantaa R, Scheinin M, Manner T,
Olkkola KT. Pharmacokinetics of intravenous dexmedetomi-
dine in children under 11 yr of age. Br J Anaesth 2008;100:
697–700
35. Diaz SM, Rodarte A, Foley J, Capparelli EV. Pharmacokinetics
of dexmedetomidine in postsurgical pediatric intensive care
unit patients: preliminary study. Pediatr Crit Care Med
2007;8:419 –24
36. Anderson PAW Developmental cardiac physiology and myo-
cardial function. In: Moller JH, Hoffman JIE, eds. Pediatric
Cardiovascular Med. New York: Churchill Livingstone, 2000:
35–55
37. Jooste EH, Muhly WT, Ibinson JW, Suresh T, Damian D,
Phadke A, Callahan P, Miller S, Feingold B, Lichtenstein SE,
Cain JG, Chrysostomou C, Davis PJ. Acute hemodynamic
changes after rapid intravenous bolus dosing of dexmedeto-
midine in pediatric heart transplant patients undergoing rou-
tine cardiac catheterization. Anesth Analg 2010;111:1490 – 6
38. Iwamoto Y, Tamai A, Kohno K, Masutani S, Okada N, Senzaki
H. Usefulness of respiratory variation of inferior vena cava
diameter for estimation of elevated central venous pressure in
children with cardiovascular disease. Circ J 2011;75:1209 –14
www.anesthesia-analgesia.org 199