ORIGINAL CONTRIBUTION
Intraoperative High-Dose Dexamethasone
for Cardiac Surgery
A Randomized Controlled Trial
Jan M. Dieleman, MD
Arno P. Nierich, MD
Peter M. Rosseel, MD
Joost M. van der Maaten, MD
Jan Hofland, MD
Jan C. Diephuis, MD
Ronald M. Schepp, MD
Christa Boer, PhD
Karel G. Moons, PhD
Lex A. van Herwerden, MD
Jan G. Tijssen, MD
Sandra C. Numan, MSc
Cor J. Kalkman, MD
Diederik van Dijk, MD
for the Dexamethasone for Cardiac
Surgery (DECS) Study Group
C
ARDIAC SURGERY IS AMONG THE
most commonly performed
surgical procedures.1 Despite
important improvements in
surgical technique, anesthesia manage-
ment, and postoperative care, cardiac sur-
gery is still associated with a substantial
risk of major adverse events.1-3
Cardiopulmonary bypass (CPB) may
play a role in the development of many
of these adverse outcomes.4,5 Cardio-
pulmonary bypass induces a complex
acute phase response, characterized by
both cell and protein activation, which
may further be intensified by the sur-
gical trauma, ischemia-reperfusion in-
jury, and endotoxemia. In a signifi-
cant number of patients, a postoperative
systemic inflammatory response syn-
drome develops, characterized by fe-
ver, organ dysfunction, and even mul-
2012 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ by a Florida State University User on 09/09/2015
Context Prophylactic corticosteroids are often administered during cardiac surgery to at-
tenuate the inflammatory response to cardiopulmonary bypass and surgical trauma; how-
ever, evidence that routine corticosteroid use can prevent major adverse events is lacking.
Objective To quantify the effect of intraoperative high-dose dexamethasone on the
incidence of major adverse events in patients undergoing cardiac surgery.
Design, Setting, and Participants A multicenter, randomized, double-blind, placebo-
controlled trial of 4494 patients aged 18 years or older undergoing cardiac surgery
with cardiopulmonary bypass at 8 cardiac surgical centers in the Netherlands enrolled
between April 13, 2006, and November 23, 2011.
Intervention Patients were randomly assigned to receive a single intraoperative dose
of 1 mg/kg dexamethasone (n=2239) or placebo (n=2255).
Main Outcome Measures A composite of death, myocardial infarction, stroke,
renal failure, or respiratory failure, within 30 days of randomization.
Results Of the 4494 patients who underwent randomization, 4482 (99.7%) could
be evaluated for the primary outcome. A total of 157 patients (7.0%) in the dexa-
methasone group and 191 patients (8.5%) in the placebo group reached the primary
study end point (relative risk, 0.83; 95% CI, 0.67-1.01; absolute risk reduction, 1.5%;
95% CI, 3.0% to 0.1%; P=.07). Dexamethasone was associated with reductions in
postoperative infection, duration of postoperative mechanical ventilation, and lengths
of intensive care unit and hospital stays. In contrast, dexamethasone was associated
with higher postoperative glucose levels.
Conclusion In our trial of adults undergoing cardiac surgery, the use of intraopera-
tive dexamethasone did not reduce the 30-day incidence of major adverse events com-
pared with placebo.
Trial Registration clinicaltrials.gov Identifier: NCT00293592
JAMA. 2012;308(17):1761-1767 www.jama.com
tisystem organ failure. It therefore dose of intravenous methylpredniso-
seems reasonable to try to attenuate the lone or dexamethasone during the sur-
inflammatory response, which is part gery. These drugs are low-cost, potent
of routine care in many cardiac surgi- anti-inflammatory agents and there-
cal centers. Often, this is accom- fore represent an appealing treatment
plished with long-acting corticoste- option in this scenario.6,7 However, con-
roids, typically by administering a high cerns about potential adverse effects, in-
Author Affiliations: Division of Anesthesiology, In- Center, Amsterdam (Dr Boer); and Academic Medi-
tensive Care, and Emergency Medicine, University cal Center, University of Amsterdam, Amsterdam (Dr
Medical Center, Utrecht (Drs Dieleman, Moons, van Tijssen), the Netherlands.
Herwerden, Kalkman, and van Dijk and Ms Numan); Members of the Dexamethasone for Cardiac Surgery
Isala Klinieken, Zwolle (Dr Nierich); Amphia Zieken- (DECS) Study Group are listed at the end of this article.
huis, Breda (Dr Rosseel); University Medical Center, Corresponding Author: Jan M. Dieleman, MD, Divi-
Groningen (Dr van der Maaten); Erasmus Medical Cen- sion of Anesthesiology, Intensive Care, and Emer-
ter, Rotterdam (Dr Hofland); Medisch Spectrum gency Medicine, University Medical Center Utrecht,
Twente, Enschede (Dr Diephuis); Medical Center, PO Box 85500, 3508 GA, Utrecht, the Netherlands
Leeuwarden (Dr Schepp); Vrije Universiteit Medical (s.dieleman@umcutrecht.nl.).
JAMA, November 7, 2012Vol 308, No. 17 1761
 INTRAOPERATIVE HIGH-DOSE DEXAMETHASONE FOR CARDIAC SURGERY
cluding inadequate serum glucose con-
trol, infectious complications, poor
wound healing, and gastrointestinal
bleeding, have precluded their wide-
spread adoption in cardiac surgical
practice.8,9 In contrast with several Eu-
ropean countries, corticosteroids are not
routinely used during cardiac surgery
in most centers in the United States.10
Previous studies have shown that cor-
ticosteroids attenuate the increase of se-
rum inflammatory markers and may im-
provepulmonarygasexchangeandreduce
the need for postoperative inotropic sup-
port.7 However, appropriately sized stud-
ies on important clinical outcomes are
lacking.Also,recentmeta-analysesdidnot
generate sufficient statistical power to al-
low conclusions on the effectiveness of
corticosteroids in the reduction of major
adverse events.11-13 As a result, corticoste-
roid administration during cardiac sur-
gery is still controversialin many Eu-
ropean hospitals it is part of routine care,
whereas this is not the case in most North
American cardiac surgical centers.14
We conducted a large randomized
clinical trial to quantify the effect of a
single intraoperative dose of dexameth-
asone on the incidence of major ad-
verse events in patients undergoing car-
diac surgery.
METHODS
Study Design and Participants
The Dexamethasone for Cardiac Sur-
gery (DECS) study is a multicenter,
randomized, double-blind, placebo-
controlled study comparing high-
dose intravenous dexamethasone with
placebo treatment in patients under-
going cardiac surgery. Between April 13,
2006, and November 23, 2011, we re-
cruited patients in 8 cardiac surgical
centers in the Netherlands. Patients
aged 18 years or older who were sched-
uled for any type of elective or urgent
cardiac surgical procedure requiring
CPB were considered eligible. Exclu-
sion criteria included an emergent or
off-pump procedure and a life expec-
tancy of less than 6 months.
The study protocol was designed by
the academic authors, in collaboration
with the members of the DECS Study
1762 JAMA, November 7, 2012Vol 308, No. 17
Downloaded From: http://jama.jamanetwork.com/ by a Florida State University User on 09/09/2015
Group. The study was conducted in ac-
cordance with Good Clinical Practice
principles and applicable national regu-
lations. The research ethics committee
at each participating center approved the
protocol. All patients were required to
provide written informed consent be-
fore randomization.
Randomization and Masking
Afterprovidingwritteninformedconsent,
patientswererandomizedtoreceiveeither
dexamethasone or placebo treatment.
Dexamethasone(1mg/kgofbodyweight,
with a 100 mg maximum) or placebo was
administered as a single intravenous in-
jection after induction of anesthesia, but
before initiation of CPB. The study drug
was supplied in packaged ampoules, each
assigned to a unique study number. Pack-
agesandampoulesofdexamethasoneand
placebo were identical and contained an
equalvolume(5mL)ofa20mg/mLdexa-
methasone solution or normal saline,
respectively. An independent statistician
created a computer-generated 1:1 ran-
domization scheme, which was stratified
toparticipatingcenterandinblocksof40.
The research pharmacist of the Univer-
sity Medical Center Utrecht, Utrecht, the
Netherlands, prepared and delivered
batches of 40 ampoules to each center.
When a consenting patient arrived in the
operating department, a packaged am-
poulewastakenfromthebatch.Whenthe
ampoule had been opened and the study
drug was administered, the patient was
considered randomized and the corre-
sponding study number was assigned to
that patient. Patients, caregivers, and re-
searchers were unaware of study group
assignment.
Procedures
Anesthesia and surgical treatment were
performed according to the standard pro-
cedures of each participating center. Sur-
gical access to the heart was achieved via
midline sternotomy. The anesthetic tech-
nique was based on either total intrave-
nous anesthesia or a combination of in-
travenous opioids and muscle relaxants
in combination with volatile anesthetics.
Techniques for cardioplegia, myocardial
protection, and CPB, as well as use of ino-
2012 American Medical Association. All rights reserved.
tropic support, antifibrinolytic therapy,
and cell saving techniques, were left at the
discretion of the attending team. Use of
corticosteroid-containing solutions for
cardioplegia or bypass circuit prime was
notallowed.Whenindicated,patientstak-
ing preoperative systemic corticosteroids
received a perioperative corticosteroid
stress regimen.
Aftersurgery,patientsweretransferred
to the intensive care unit (ICU) and
weanedfrommechanicalventilationwhen
therewasnoexcessiveongoingbloodloss
and patients were cooperative and hemo-
dynamically stable. Perioperative serum
glucose was regulated according to local
sliding scale protocols.
Statistical Analysis
The primary study end point of major ad-
verse events was a composite of death,
myocardial infarction (MI), stroke, renal
failure, or respiratory failure, occurring
within 30 days of randomization. Peri-
operative MI was defined as the presence
of new Q waves or a new left bundle
branch block on the electrocardiogram,
combined with a biomarker (creatine
kinaseMBortroponin)elevationofmore
than 5 times the upper reference limit.
Data from routine cardiac biomarker
surveillance were used to detect possible
perioperative MI. The specific type of bio-
markerusedwasdictatedbythelocalpro-
tocol in each center, rather than by the
study protocol. Postdischarge MI was de-
fined according to the criteria of the Uni-
versal Definition of Myocardial Infarc-
tion.15 Stroke was defined as a neurologic
deficit lasting more than 24 hours, with
increased invalidity (increase on Rankin
scale16 of 1 point) and signs of a new
ischemic cerebral infarction on com-
puted tomography or magnetic resonance
imaging scan. Renal failure in patients not
previously receiving dialysis was defined
according to the RIFLE criteria as an in-
crease in postoperative serum creatinine
of at least 3 times the preoperative value,
or a serum creatinine level of more than
4 mg/dL (to convert to micromoles per
liter, multiply by 88.4) associated with an
acute increase of serum creatinine of at
least 0.5 mg/dL.17 Respiratory failure was
defined as postoperative mechanical ven-

tilation or reinstitution of mechanical re-
spiratory support via an orotracheal tube
or tracheostomy for an uninterrupted pe-
riod of at least 48 hours.
An exploratory analysis of prospec-
tively defined secondary outcomes in-
cluded each separate component of the
primary end point (ie, death, MI, stroke,
renal failure, or respiratory failure, within
the first 30 days); postoperative infec-
tions; postoperative atrial fibrillation;
highest serum glucose concentration in
the ICU; highest body temperature in the
ICU; postoperative delirium (defined as
the postoperative indication for treat-
ment with neuroleptic drugs); time to
weaning from postoperative mechani-
cal ventilation; and time to discharge
from the ICU and from the hospital.
An independent, blinded critical event
adjudicationcommitteereviewedallcases
of death, possible MI, and possible stroke.
Cases of possible MI or stroke were either
confirmed or revoked according to the
study definitions of these events.
We hypothesized that dexamethasone
administration would reduce the inci-
denceoftheprimarystudyoutcome.Based
on the Society of Thoracic Surgeons da-
tabase,18 the incidence of the primary out-
come in the placebo group was estimated
to be 6%. To detect an absolute difference
of 2% (from 6% to 4%) with a power of
80% at a 2-sided .05 significance level,
1962 patients would be required in each
study group. To compensate for possible
lossestofollow-up,weplannedtoinclude
2250 patients per study group.
During the study, 3 preplanned in-
terim analyses on the primary study end
point were performed when 1000, 2000,
and 3250 patients, respectively, had been
enrolled. Interim analyses were per-
formed by the independent data and
safety monitoring board, which con-
sisted of an epidemiologist, a cardiac sur-
geon, and a cardiac anesthesiologist not
involved in the study. These blinded
analyses were adjusted according to an
OBrien and Fleming type I error spend-
ing function,19 using an overall .05 sig-
nificance level. As a result of these in-
terim analyses, the threshold for
significance of the primary study end
point in the final analysis was .044.
2012 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ by a Florida State University User on 09/09/2015
INTRAOPERATIVE HIGH-DOSE DEXAMETHASONE FOR CARDIAC SURGERY
Figure 1. Enrollment Flowchart
25 085 Patients assessed for eligibility
3504 Not eligible for inclusion because of planned
off-pump surgery
21 581 Eligible for inclusion
17 087 Excluded
16 754 Not invited or declined participation
333 Provided informed consent but not
randomized
14 Informed consent withdrawn
25 Corticosteroids considered indicated
or contraindicated a
65 Not randomized for logistic reasons
114 Intervention canceled
115 Other reasons
4494 Randomized
2239 Randomized to receive dexamethasone 2255 Randomized to receive placebo
2239 Received dexamethasone as 2255 Received placebo as randomized
randomized
4 No follow-up for primary end point 8 No follow-up for primary end point
2 Withdrew informed consent 5 Withdrew informed consent
2 Patients could not be traced 3 Patients could not be traced
2235 Included in primary analysis 2247 Included in primary analysis
4 Excluded due to no follow-up 8 Excluded due to no follow-up
Data on the number of patients that were not invited to participate, or that declined participation, were not con-
sistently logged in all centers and are therefore not sufficiently accurate to be reported in detail. aIndication or
contraindication dictated by either the treating medical team or the clinical situation during the start of the pro-
cedure in the operating department.
Patient follow-up for secondary out- ondary outcomes, we used the 2 test.
comes was until 1 year from randomiza- Absolute risk reduction or relative risk
tion by study protocol. Herein, we report (RR) with 95% CIs was calculated for
theprimarystudyendpointtogetherwith each dichotomous outcome measure.
exploratory analyses of other outcomes Logistic regression was used for assess-
in the first 30 days after randomization. ing heterogeneity in the subgroup
Analyses were conducted according to analyses, with a .10 threshold for sig-
randomization (intention-to-treat analy- nificance. For comparison of mean and
ses).Baselinecharacteristicsinthe2study median values of the continuous sec-
groups were evaluated using frequency ondary outcome measures, we used Stu-
distributions. dent t test or Mann-Whitney U test, as
We also performed preplanned sub- appropriate. IBM SPSS version 19 (SPSS
group analyses for the primary out- Inc) was used for all analyses.
come and its separate components,
which included 4 age groups (65, 65- RESULTS
74, 75-79, and 80 years), sex, diabe- Study Population
tes, chronic obstructive pulmonary dis- An estimated 25 085 patients sched-
ease, higher (5) vs low (4) EuroScore uled to undergo elective or urgent car-
preoperative risk estimate20 (cutoff value diac surgery were screened, of whom
based on the median EuroScore of the 21 581 were eligible for inclusion. Of the
study population), and prolonged CPB 4827 patients who provided written in-
duration (defined as 150 minutes). formed consent, 4494 eventually under-
For the comparison of the propor- went randomization (FIGURE 1). Two
tions of patients with primary and sec- patients who were unintentionally
JAMA, November 7, 2012Vol 308, No. 17 1763
 INTRAOPERATIVE HIGH-DOSE DEXAMETHASONE FOR CARDIAC SURGERY

randomized without having provided the dexamethasone group and 5 sone group and 3 patients (0.1%) in
informed consent were excluded patients (0.2%) in the placebo group the placebo group after hospital dis-
from the analysis. Of the 4494 ran- withdrew consent; therefore, their charge. Thus, the analyzed popula-
domized patients, 2239 (49.8%) were primary outcome could not be tion consisted of 4482 patients
randomized to dexamethasone treat- assessed. We were unable to obtain (99.7%). Patients in the study groups
ment and 2255 (50.2%) to placebo 30-day outcome information in 2 were similar with respect to baseline
treatment. Two patients (0.1%) in patients (0.1%) in the dexametha- demographic, clinical, and surgical
characteristics (TABLE 1).
Table 1. Demographic, Clinical, and Surgical Characteristics of the Dexamethasone and Primary Study End Point
Placebo Groups a
Dexamethasone Placebo In total, 348 patients (7.8%) reached the
Characteristics (n = 2235) (n = 2247) composite primary study end point of
Demographics death, MI, stroke, renal failure, or re-
Age, mean (SD), y 66.2 (11.0) 66.1 (10.7) spiratory failure, within 30 days after
Male sex 1622 (72.6) 1628 (72.5)
randomization (TABLE 2). The pri-
Weight, mean (SD), kg 82.4 (14.3) 82.0 (14.4)
mary study end point occurred in 157
Height, mean (SD), cm 174 (9.1) 173 (9.2)
of the 2235 patients (7.0%) random-
Coexisting medical conditions
Hypertension 1179 (54.7) 1180 (54.8) ized to dexamethasone and in 191 of
Diabetes mellitus the 2247 patients (8.5%) randomized
Insulin dependent 106 (4.8) 125 (5.8) to placebo (RR, 0.83; 95% CI, 0.67-
Noninsulin dependent 309 (13.9) 311 (13.9) 1.01; absolute risk reduction, 1.5%;
Treatment for pulmonary disease 243 (10.9) 266 (11.9) 95% CI, 3.0% to 0.1%; P =.07).
Previous cerebrovascular event
Stroke 86 (3.9) 78 (3.5) Exploratory Analysis of the
Transient ischemic attack 107 (4.8) 103 (4.6) Combined End Point Components
Peripheral vascular disease 191 (8.6) 192 (8.6)
The rate of death, MI, stroke, and re-
Preoperative creatinine, mean (SD), mg/dL 1.04 (0.38) 1.07 (0.57)
nal failure was similar in both groups.
Chronic renal dysfunction 17 (0.8) 29 (1.3)
In the dexamethasone group, 67 pa-
Cardiac status
Recent myocardial infarction (90 d) 195 (8.7) 176 (7.8) tients (3.0%) experienced respiratory
Left ventricular function b failure compared with 97 patients
Moderate 503 (22.6) 534 (23.9) (4.3%) in the placebo group (RR, 0.69;
Poor 103 (4.6) 117 (5.2) 95% CI, 0.51-0.94; P =.02) (Table 2).
EuroScore, median (IQR) c 5 (3-7) 5 (3-7) The RR for a composite end point with-
Preoperative medication out the respiratory failure compo-
-Blocker 1485 (68.4) 1479 (68.2)
nent, consisting of only mortality, MI,
Statin 836 (58.0) 771 (53.8)
stroke, and renal failure, was 0.84 (95%
Corticosteroid 130 (7.2) 98 (5.4)
Type of surgery
CI, 0.66-1.08; P=.18).
Isolated CABG 873 (39.1) 895 (39.8)
CABG plus valve 360 (16.1) 388 (17.3) Secondary End Points
Single valve 575 (25.7) 558 (24.8) The median (interquartile range [IQR])
Surgery on multiple valves 82 (3.7) 99 (4.4) time to weaning from mechanical ven-
Other procedures 336 (15.0) 295 (13.1) tilation was 7.0 (4.7-10.0) hours (mean,
Repeat surgery 140 (6.3) 147 (6.6) 11.0 hours) in the dexamethasone group
Duration of procedure, mean (SD), min 244 (102) 242 (93) and 7.0 (5.0-11.0) hours (mean, 14.3
Duration of extracorporeal circulation, mean (SD), min 125 (68) 124 (64) hours) in the placebo group (P.001)
Duration of aortic cross-clamping, mean (SD), min 87 (47) 85 (44) (TABLE 3). The median (IQR) time to dis-
Deep hypothermic circulatory arrest 15 (0.7) 23 (1.0) charge from the ICU was 22.0 (19.0-
Use of cell-saving device 1151 (51.8) 1104 (49.4) 24.0) hours (mean, 34.2 hours) in the
Use of antifibrinolytic drug dexamethasone group and 22.0 (19.0-
Tranexamic acid 1834 (82.4) 1835 (81.8)
25.0) hours (mean, 43.6 hours) in the
Other antifibrinolytic drug 10 (0.4) 18 (0.8)
placebo group (P.001). The low P val-
Abbreviations: CABG, coronary artery bypass graft; IQR, interquartile range.
SI conversion: To convert creatinine to mol/L, multiply by 88.4. ues for both comparisons despite simi-
a Data are shown as No. (%) unless otherwise indicated.
b Definition of left ventricular function classes20: moderate, ejection fraction of 30% to 50%; and poor, ejection fraction lar median values are the result of a
of less than 30%.
c Higher EuroScores present increased risk of perioperative mortality.20
higher proportion of patients requiring
prolongedventilationtimesandprolonged
1764 JAMA, November 7, 2012Vol 308, No. 17 2012 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Florida State University User on 09/09/2015

 INTRAOPERATIVE HIGH-DOSE DEXAMETHASONE FOR CARDIAC SURGERY

ICU stay in the placebo group. For the mortality component of the pri- incidence of the composite primary
example, the proportion of patients mary study end point, which showed study end point of major adverse
requiring more than 24 hours of post- significant heterogeneity (P=.05). In pa- events (P = .07). In an exploratory
operative mechanical ventilation was tients younger than 65 years, the RR analysis of secondary end points, a
3.4% in the dexamethasone group for mortality was 0.42 (95% CI, 0.13- reduced incidence of respiratory fail-
compared with 4.9% in the placebo 1.34; P = .13), but it gradually in- ure was found, which was accompa-
group. Similarly, the proportion of creased with age to 3.87 (95% CI, 1.10- nied by an overall reduced time to
patients requiring more than 48 hours 13.6; P =.02) in patients aged 80 years weaning from mechanical ventila-
of ICU stay was lower in the dexa- or older. tion, a lower risk of pneumonia, and
methasone group than in the placebo a reduction in ICU and hospital stay.
group (10.2% vs 14.0%, respectively). COMMENT The DECS trial is the first large ran-
The median (IQR) time to discharge Our randomized study of 4494 domized controlled trial to our knowl-
from the hospital in the dexametha- patients undergoing cardiac surgery edge on the controversial topic of rou-
sone group was 8 (7-13) days (mean, failed to show a statistically signifi- tine corticosteroid use during cardiac
11.3 days) vs 9 (7-13) days (mean, cant benefit of intraoperative admin- surgery in adults. The numerous small
11.7 days) in the placebo group istration of dexamethasone on the randomized studies published in the last
(P=.009).
The risk of developing a postopera-
Table 2. Primary Study End Point and Components of the Primary Study End Point in the
tive infection was lower in the dexa- Dexamethasone and Placebo Groups
methasone group than in the placebo No. (%) of Patients
group (9.5% vs 14.8%, respectively; RR,
0.64; 95% CI, 0.54-0.75; P .001) Dexamethasone Placebo Relative Risk
(n = 2235) (n = 2247) (95% CI)
(Table 3). This protective effect was pri- Primary study end point a 157 (7.0) 191 (8.5) 0.83 (0.67-1.01)
marily related to a decreased inci- Components of the primary study end point
dence of pneumonia in the dexameth- Death 31 (1.4) 34 (1.5) 0.92 (0.57-1.49)
asone group (6.0% vs 10.6% in the Myocardial infarction 35 (1.6) 39 (1.7) 0.90 (0.57-1.42)
placebo group; RR, 0.56; 95% CI, 0.46- Stroke 29 (1.3) 32 (1.4) 0.91 (0.55-1.50)
0.69; P.001). In contrast, the mean Renal failure 28 (1.3) 40 (1.8) 0.70 (0.44-1.14)
highest serum glucose concentration Respiratory failure 67 (3.0) 97 (4.3) 0.69 (0.51-0.94)
was higher in the dexamethasone group a Primary study end point was a composite of death, myocardial infarction, stroke, renal failure, or respiratory failure,
within 30 days after surgery.
and the incidence of postoperative fe-
ver was higher in the placebo group
(Table 3).
Table 3. Secondary End Points in the Dexamethasone and Placebo Groups
Subgroup Analysis Dexamethasone Placebo Relative Risk P
Secondary End Points (n = 2235) (n = 2247) (95% CI) Value a
The preplanned subgroup analyses Median (IQR)
suggested an age-dependent effect of Duration of postoperative mechanical 7.0 (4.7-10.0) 7.0 (5.0-11.0) NA .001
ventilation, h
dexamethasone on the primary study
Length of stay in the ICU, h 22.0 (19.0-24.0) 22.0 (19.0-25.0) NA .001
end point (P for heterogeneity = .08)
Length of hospital stay, d 8 (7-13) 9 (7-13) NA .009
(FIGURE 2). In patients younger than
Mean (SD)
65 years, dexamethasone was associ- Highest serum glucose concentration 195 (50) 177 (59) NA .001
ated with lower likelihood for the in the ICU, mg/dL
primary end point (RR, 0.65; 95% No. (%)
Body temperature in the ICU of 38C 132 (5.9) 564 (25.2) 0.23 (0.20-0.27) .001
CI, 0.44-0.96; P = .03), whereas in
Atrial fibrillation 739 (33.1) 790 (35.2) 0.94 (0.87-1.02) .14
patients aged 80 years or older, the Delirium 205 (9.2) 262 (11.7) 0.79 (0.66-0.94) .006
RR was 1.69 (95% CI, 0.92-3.10; Gastrointestinal bleeding 13 (0.6) 11 (0.5) 1.19 (0.53-2.65) .67
P = .09). There was no differential Any postoperative infection 212 (9.5) 333 (14.8) 0.64 (0.54-0.75) .001
treatment effect in the subgroup Pneumonia 133 (6.0) 238 (10.6) 0.56 (0.46-0.69) .001
analyses on sex, diabetes, chronic Urinary tract infection 50 (2.2) 60 (2.7) 0.84 (0.58-1.21) .35
obstructive pulmonary disease, Wound infection 34 (1.5) 32 (1.4) 1.07 (0.66-1.72) .79
EuroScore, or prolonged CPB dur- Catheter-related infection 6 (0.3) 21 (0.9) 0.29 (0.12-0.71) .004
ation. Sepsis 18 (0.8) 26 (1.2) 0.70 (0.38-1.27) .23
The bidirectional effect of dexameth- Abbreviations: ICU, intenstive care unit; IQR, interquartile range; NA, not applicable.
asone across the 4 age groups ap- SI conversion: To convert glucose to mmol/L, multiply by 0.0555.
a For parametric and nonparametric comparisons of continuous data.
peared to be predominantly caused by
2012 American Medical Association. All rights reserved. JAMA, November 7, 2012Vol 308, No. 17 1765

Downloaded From: http://jama.jamanetwork.com/ by a Florida State University User on 09/09/2015

 INTRAOPERATIVE HIGH-DOSE DEXAMETHASONE FOR CARDIAC SURGERY

Figure 2. Forest Plot of Subgroup Analyses

No. of Patients No. of Events

Relative Risk Dexamethasone Placebo P for
Subgroup Dexamethasone Placebo Dexamethasone Placebo (95% CI) Better Better Trend
Sex
Male 1622 1628 112 130 0.86 (0.68-1.10)
.50
Female 613 619 45 61 0.74 (0.52-1.08)
Age, y
<65 913 962 38 62 0.65 (0.44-0.96)
65-74 818 781 57 70 0.78 (0.56-1.09)
75-79 344 336 38 44 0.84 (0.56-1.27) .08
80 159 168 24 15 1.69 (0.92-3.10)
EuroScore
0-4 1025 1060 43 46 0.97 (0.64-1.45)
.32
5 1180 1147 113 142 0.77 (0.61-0.98)
Diabetes
No 1816 1805 115 142 0.81 (0.64-1.02)
.73
Yes 415 436 41 49 0.88 (0.59-1.30)
COPD
No 1987 1978 133 153 0.87 (0.69-1.08)
.39
Yes 243 266 24 38 0.69 (0.43-1.12)
Prolonged CPB
No 1674 1659 85 92 0.92 (0.69-1.22)
.37
Yes 536 563 71 96 0.78 (0.59-1.03)
Total 2235 2247 157 191 0.83 (0.67-1.01)

0.2 1.0 5.0

Relative Risk (95% CI)

COPD indicates chronic obstructive pulmonary disease; CPB, cardiopulmonary bypass. The effect estimates for the primary study end point in the subgroup analyses
are shown. The size of each data marker correlates with the total number of patients in that subgroup.

decades had conflicting results13 and
provided only limited guidance for se-
lection of components for the compos-
ite primary end point of our study. No
significant benefit from dexametha-
sone treatment was observed on the
composite primary end point, which
was largely cardiovascular. However,
further exploration of the study data
suggested a consistent pattern of im-
proved pulmonary condition, mani-
fested as a lower risk of postoperative
respiratory failure, shorter times to
weaning from the ventilator, and re-
duced risk of pneumonia during post-
operative hospitalization in the dexa-
methasone group. This improved
respiratory condition was accompa-
nied by earlier discharge from both the
ICU and the hospital. It might be ar-
gued that the beneficial effect of dexa-
methasone on these multiple second-
ary outcomes was only a coincident
finding. However, also at a more con-
servative cutoff value for statistical sig-
nificance of .0025to correct for test-
ing a total of 19 secondary outcomes
using a Bonferroni approachmost of
the effects found remained signifi-
cant. These effects may actually be in-
1766 JAMA, November 7, 2012Vol 308, No. 17
terrelated and explained by attenua-
tion of the perioperative systemic
inflammatory response syndrome,
which is the pathophysiologic goal of
the single high-dose corticosteroid
treatment.11-14 Thus, although the ef-
fect of dexamethasone on the primary
end point was negative, there is the pos-
sibility that a clinically significant ef-
fect was missed. Therefore, a new pro-
spective study focusing on pulmonary
outcomes seems a logical next step to
further explore the secondary find-
ings of our trial. Such a study should
also consider patient selection for the
therapy as we found a larger beneficial
effect in younger patients and no ap-
parent benefit in those aged 80 years or
older.
Our study failed to confirm that cor-
ticosteroids reduce the incidence of
postoperative atrial fibrillation, as dem-
onstrated in a previous study.21 The 241
patients in that particular trial re-
ceived moderate-dose hydrocortisone
3 days postoperatively, instead of the
single intraoperative high dose of dexa-
methasone in our study.
The reduced risk of postoperative in-
fections, in particular pulmonary in-
2012 American Medical Association. All rights reserved.
fections, in the dexamethasone group
was unexpected, and contrary to exist-
ing knowledge that corticosteroids in-
crease the risk of infections.7 How-
ever, this adverse effect is mainly related
to chronic corticosteroid use, rather
than to a single prophylactic pulse dose
in circumstances wherein the activa-
tion of the immune system could be det-
rimental.11-14 The reduced infection risk
persisted when patients with respira-
tory failure were excluded from the
analysis, suggesting that the observa-
tion is not the result of shorter expo-
sure to mechanical ventilation.
A pooled analysis of this study with
the results of our previous meta-
analysis of 56 studies on prophylactic
corticosteroids in cardiac surgery,
which was recently published,13 showed
no effect of corticosteroids for mortal-
ity or cardiac complications. How-
ever, corticosteroid treatment re-
duced respiratory complications (Peto
odds ratio, 0.59; 95% CI, 0.49-0.71;
P.001) (eMethods and eFigure, avail-
able at http://www.jama.com).
A limitation of our trial is that we
studied a high-dose dexamethasone
regimen, which is often used during

Downloaded From: http://jama.jamanetwork.com/ by a Florida State University User on 09/09/2015


cardiac surgery in several European
countries. In North America, however,
methylprednisolone is usually pre-
ferred when corticosteroids are admin-
istered during cardiac surgery. The ef-
fect of high-dose methylprednisolone for
cardiac surgery is being studied in an on-
going large study (Steroids In cardiac
Surgery [SIRS] trial, NCT00427388). A
strength of our study is that blinding for
treatment was well maintained during
the perioperative period. The small dif-
ferences in serum glucose and postop-
erative body temperature are unlikely to
have caused awareness of randomiza-
tion. It is thus unlikely that such aware-
ness, if any, could have influenced clini-
cal management to a degree that could
have produced the present effects on pul-
monary outcome and duration of ICU
and hospital stay.
In conclusion, in our trial of adults un-
dergoing cardiac surgery, the use of in-
traoperative dexamethasone did not re-
duce the 30-day incidence of major
adverse events compared with placebo.
REFERENCES
1. Roger VL, Go AS, Lloyd-Jones DM, et al; Ameri-
can Heart Association Statistics Committee and Stroke
Statistics Subcommittee. Heart disease and stroke sta-
tistics--2011 update: a report from the American Heart
Association. Circulation. 2011;123(4):e18-e209.
2. Newman MF, Mathew JP, Grocott HP, et al. Cen-
tral nervous system injury associated with cardiac
surgery. Lancet. 2006;368(9536):694-703.
3. Mariscalco G, Lorusso R, Dominici C, Renzulli A,
Sala A. Acute kidney injury: a relevant complication
after cardiac surgery. Ann Thorac Surg. 2011;92
(4):1539-1547.
4. Moat NE, Shore DF, Evans TW. Organ dysfunc-
tion and cardiopulmonary bypass: the role of comple-
ment and complement regulatory proteins. Eur J Car-
diothorac Surg. 1993;7(11):563-573.
5. Asimakopoulos G. Systemic inflammation and car-
diac surgery: an update. Perfusion. 2001;16(5):
353-360.
6. Wan S, LeClerc JL, Vincent JL. Inflammatory re-
sponse to cardiopulmonary bypass: mechanisms in-
volved and possible therapeutic strategies. Chest. 1997;
112(3):676-692.
7. Chaney MA. Corticosteroids and cardiopulmo-
nary bypass : a review of clinical investigations. Chest.
2002;121(3):921-931.
8. Vogelzang M, Hoekstra M, Drost JT, et al. The im-
pact of a reduced dose of dexamethasone on glu-
2012 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ by a Florida State University User on 09/09/2015
INTRAOPERATIVE HIGH-DOSE DEXAMETHASONE FOR CARDIAC SURGERY
Author Contributions: Drs Dieleman and van Dijk had
full access to all of the data in the study and take
responsibility for the integrity of the data and the
accuracy of the data analysis.
Study concept and design: Dieleman, Nierich, Rosseel,
van der Maaten, Hofland, Diephuis, Moons,
van Herwerden, Tijssen, Kalkman, van Dijk.
Acquisition of data: Dieleman, Nierich, Rosseel,
van der Maaten, Hofland, Diephuis, Schepp, Boer,
van Herwerden, Numan.
Analysis and interpretation of data: Dieleman,
van der Maaten, Hofland, Schepp, Moons, Tijssen,
Kalkman, van Dijk.
Drafting of the manuscript: Dieleman, Moons, Tijssen,
Kalkman, van Dijk.
Critical revision of the manuscript for important in-
tellectual content: Dieleman, Nierich, Rosseel,
van der Maaten, Hofland, Diephuis, Schepp, Boer,
Moons, van Herwerden, Tijssen, Numan, Kalkman,
van Dijk.
Statistical analysis: Dieleman, Moons, Tijssen, Kalkman,
van Dijk.
Obtained funding: Dieleman, Moons, Kalkman,
van Dijk.
Administrative, technical, or material support:
Dieleman, Nierich, Numan.
Study supervision: Dieleman, Moons, Kalkman,
van Dijk.
Conflict of Interest Disclosures: All authors have com-
pleted and submitted the ICMJE Form for Disclosure
of Potential Conflicts of Interest and none were re-
ported.
Funding/Support: This work was supported by
grants 80-82310-98-08607 from the Netherlands
Organization for Health Research and Development
(ZonMw) and 2007B125 from the Dutch Heart
Foundation.
cose control after coronary artery bypass surgery. Car-
diovasc Diabetol. 2007;6(1):39.
9. Duncan AE, Abd-Elsayed A, Maheshwari A, Xu M,
Soltesz E, Koch CG. Role of intraoperative and post-
operative blood glucose concentrations in predicting
outcomes after cardiac surgery. Anesthesiology. 2010;
112(4):860-871.
10. Sulzer CF, Mackensen GB, Grocott HP. Con: Meth-
ylprednisolone is not indicated for patients during car-
diopulmonary bypass. J Cardiothorac Vasc Anesth.
2005;19(2):255-258.
11. Whitlock RP, Chan S, Devereaux PJ, et al. Clini-
cal benefit of steroid use in patients undergoing car-
diopulmonary bypass: a meta-analysis of random-
ized trials. Eur Heart J. 2008;29(21):2592-2600.
12. Ho KM, Tan JA. Benefits and risks of corticoste-
roid prophylaxis in adult cardiac surgery: a dose-
response meta-analysis. Circulation. 2009;119
(14):1853-1866.
13. Dieleman JM, van Paassen J, van Dijk D, et al. Pro-
phylactic corticosteroids for cardiopulmonary bypass
in adults. Cochrane Database Syst Rev. 2011;(5):
CD005566.
14. Whitlock RP, Rubens FD, Young E, Teoh KH. Pro:
Steroids should be used for cardiopulmonary bypass.
J Cardiothorac Vasc Anesth. 2005;19(2):250-254.
15. Thygesen K, Alpert JS, White HD; Joint ESC/
ACCF/AHA/WHF Task Force for the Redefinition of
JAMA, November 7, 2012Vol 308, No. 17 1767
Role of the Sponsors: The sponsors had no role in the
design and conduct of the study; in the collection,
analysis, and interpretation of the data; or in the prepa-
ration, review, or approval of the manuscript.
Members of the DECS Study Group (the Nether-
lands): University Medical Center, Utrecht: Jaap J. Bre-
dee, MD, Wolfgang F. Buhre, MD, Jan M. Dieleman,
MD, Diederik van Dijk, MD, Lex A. van Herwerden,
MD, Cor J. Kalkman, MD, Jan van Klarenbosch, MD,
Karel G. Moons, PhD, Hendrik M. Nathoe, MD, San-
dra C. Numan, MSc, Thomas H. Ottens, MD, Kit C.
Roes, PhD, Anne-Mette C. Sauer, MD, Arjen J. Slooter,
MD; Isala Klinieken, Zwolle: Arno P. Nierich, MD, Ja-
cob J. Ennema, MD; Amphia Ziekenhuis, Breda: Pe-
ter M. Rosseel, MD, Nardo J. van der Meer, MD; Uni-
versity Medical Center, Groningen: Joost M. van der
Maaten, MD, Vlado Cernak, MD; Erasmus Medical
Center, Rotterdam: Jan Hofland, MD, Robert J. van
Thiel, MD; Medisch Spectrum Twente, Enschede: Jan
C. Diephuis, MD; Medical Center, Leeuwarden: Ron-
ald M. Schepp, MD, Jo Haenen, MD, Fellery de Lange,
MD; Vrije Universiteit Medical Center, Amsterdam:
Christa Boer, PhD, Jan R. de Jong, MD; Academic
Medical Center, Amsterdam: Jan G. Tijssen, MD.
Steering Committee: Jan M. Dieleman, MD, Jan C.
Diephuis, MD, Diederik van Dijk, MD, Lex A. van Her-
werden, MD, Jan Hofland, MD, Jan R. de Jong, MD,
Cor J. Kalkman, MD, Jan van Klarenbosch, MD, Joost
M. van der Maaten, MD, Karel G. Moons, PhD, Arno
P. Nierich, MD, Peter M. Rosseel, MD, Ronald M.
Schepp, MD.
Data and Safety Monitoring Board: Peter Bruins, MD,
Bas A. de Mol, MD, Jan G. Tijssen, MD.
Critical Event Adjudication Committee: Jaap J. Bredee,
MD, Hendrik M. Nathoe, MD, Arjen J. Slooter, MD.
Online-Only Material: eMethods and eFigure are avail-
able at http://www.jama.com.
Myocardial Infarction. Universal definition of myo-
cardial infarction. Eur Heart J. 2007;28(20):2525-
2538.
16. van Swieten JC, Koudstaal PJ, Visser MC, Schouten
HJ, van Gijn J. Interobserver agreement for the as-
sessment of handicap in stroke patients. Stroke. 1988;
19(5):604-607.
17. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky
P; Acute Dialysis Quality Initiative workgroup. Acute
renal failure: definition, outcome measures, animal
models, fluid therapy and information technology
needs: the Second International Consensus Confer-
ence of the Acute Dialysis Quality Initiative (ADQI)
Group. Crit Care. 2004;8(4):R204-R212.
18. The Society of Thoracic Surgeons National
Database. http://www.ctsnet.org/section
/stsdatabase. Accessed October 1, 2012.
19. OBrien PC, Fleming TR. A multiple testing pro-
cedure for clinical trials. Biometrics. 1979;35(3):
549-556.
20. Nashef SA, Roques F, Michel P, Gauducheau E,
Lemeshow S, Salamon R. European system for car-
diac operative risk evaluation (EuroSCORE). Eur J Car-
diothorac Surg. 1999;16(1):9-13.
21. Halonen J, Halonen P, Jarvinen O, et al. Cortico-
steroids for the prevention of atrial fibrillation after car-
diac surgery: a randomized controlled trial. JAMA.
2007;297(14):1562-1567.