Acute Ischemic Heart Disease

Nitrous oxide and perioperative cardiac morbidity

(ENIGMA-II) Trial: Rationale and design
Paul S. Myles, MPH, MD, a,b,c,d Kate Leslie, MEpi, MD, e Philip Peyton, MBBS, MD, f Michael Paech, MB.BS, DM, g
Andrew Forbes, MSc, PhD, c Matthew T.V. Chan, MB.BS, FANZCA, h Daniel Sessler, MD, i
Philip J. Devereaux, MD, PhD, j Brendan S. Silbert, MBBS, FANZCA, k Konrad Jamrozik, MB.BS, DPhil, l
Scott Beattie, MD, PhD, m Neal Badner, MD, FRCP(C), n James Tomlinson, MB.BS, FRACP, a,b and Sophia Wallace,
BSc a the ANZCA Trials Group Melbourne, Victoria, Perth, WA, and Adelaide SA, Australia; Hong Kong, China;
Cleveland, OH; and Hamilton, Toronto, and London, Ontario, Canada

Background Globally there are N200 million major surgical procedures undertaken annually, and about 20% of these
involve patients who have coronary artery disease. Many receive nitrous oxide, which impairs methionine synthase, thus
inhibiting folate synthesis and increasing postoperative homocysteine levels. Nitrous oxide anesthesia leads to postoperative
endothelial dysfunction, and there is some evidence that it increases myocardial ischemia and, possibly, myocardial
infarction. We have initiated the Nitrous oxide and perioperative cardiac morbidity (ENIGMA-II) Trial to test the hypothesis
that in inpatients undergoing anesthesia for major noncardiac surgery, avoidance of nitrous oxide will reduce the incidence
of death and major cardiovascular events.
Methods ENIGMA-II is a 7,000-patient, international randomized trial involving patients at risk of coronary artery
disease undergoing noncardiac surgery. The patients, health care providers (except for the anesthesiologists), data
collectors, and outcome adjudicators are blinded to whether patients receive nitrous oxidecontaining or nitrous oxidefree
anesthetic. The primary outcome is a composite of death and major nonfatal events (ie, myocardial infarction, cardiac arrest,
pulmonary embolism, and stroke) at 30 days after surgery.
Results At present, ENIGMA-II has randomized N1,000 patients in 22 hospitals in 5 countries. To date, patients' mean
age is 70 years, 66% are men, 38% have a history of coronary artery disease, 19% have a history of cerebrovascular
disease, and 84% have a history of hypertension. Most patients have undergone intra-abdominal 28%, vascular 32%, and
orthopedic 16% surgery.
Conclusions The ENIGMA-II Trial will be the largest study yet conducted to ascertain the benefits and risks of
removing nitrous oxide from the gas mixture in anesthesia. The results of this large international trial will guide the clinical
care of the hundreds of millions of adults undergoing noncardiac surgery annually. (Am Heart J 2009;157:488-494.e1)

Globally there are N230 million major surgical proce-
dures undertaken annually,1 and given that cardiac and
pediatric surgery only account for a few of major surgical
cases, this suggests that N200 million adults undergo
major noncardiac surgery annually. Millions of these
patients will have a major perioperative cardiovascular
event (ie, cardiovascular death, nonfatal myocardial
infarction [MI], nonfatal cardiac arrest, or nonfatal
stroke). These perioperative cardiovascular complica-
tions prolonged hospitalization and increase costs to the
health system.2-4 Recently, some of us undertook the
POISE (PeriOperative ISchemic Evaluation) trial,5 the
largest perioperative cardiovascular noncardiac surgery
trial conducted to date. In POISE, 6.4% of the patients had

From the a
Department of Anaesthesia and Perioperative Medicine, Alfred Hospital,
Melbourne, Victoria, Australia, bAcademic Board of Anaesthesia and Perioperative
Medicine, Monash University, Melbourne, Victoria, Australia, cDepartment of Epidemio-
logy and Preventive Medicine, Monash University; Melbourne, Victoria, Australia,
d
Australian National Health and Medical Research Council Centre for Clinical Research
Excellence in Therapeutics, Melbourne, Victoria, Australia, eDepartment of Anaesthesia
and Pain Management, Royal Melbourne Hospital, Melbourne, Victoria, Australia,
f
Department of Anaesthesia, Austin Hospital, Heidelberg, Victoria, Australia, gThe School
of Medicine and Pharmacology, The University of Western Australia, Perth, WA, Australia,
h
Department of Anaesthesia, Prince of Wales Hospital, The Chinese University of Hong
Kong, Hong Kong, China, iDepartment of Outcomes Research, The Cleveland Clinic,
Cleveland, OH, jDepartments of Clinical Epidemiology and Biostatististics and Medicine,
McMaster University, Hamilton, Ontario, Canada, kDepartment of Anaesthesia,
St Vincent's Hospital, Fitzroy, Victoria, Australia, lSchool of Population Health & Clinical
m
Practice, University of Adelaide, SA, Australia, Department of Anaesthesia, University
n
Health Network, Toronto, Ontario, Canada, and Department of Anesthesiology,
University of Western Ontario, London, Ontario, Canada.
Trial Registration: www.clinicaltrials.gov CT00430989.
For ENIGMA-II Trial Organization and a list of committees see Appendix A, available online.
Submitted September 30, 2008; accepted November 25, 2008.
Reprint requests: Paul S. Myles, MPH, MD, Department of Anaesthesia and Perioperative
Medicine, Alfred Hospital, Commercial Road, Melbourne, Victoria 3004, Australia.
E-mail: p.myles@alfred.org.au
0002-8703/$ - see front matter
2009, Mosby, Inc. All rights reserved.
doi:10.1016/j.ahj.2008.11.015
American Heart Journal
Volume 157, Number 3
a major cardiovascular event (ie, cardiovascular death,
nonfatal MI, or nonfatal cardiac arrest) within the first 30
days. A total of 2.7% of the patients died within the first 30
days, and 59% of the fatalities were adjudicated as a
cardiovascular death. A perioperative MI has been
estimated to add $15,000 to the costs of a hospital stay,
whereas a death adds an incremental cost of N$20,000.3,4
The cost of perioperative cardiac events is estimated at
$20 billion annually in the United States alone.2,6
Nitrous oxide and anesthesia
Despite some concerns regarding the safety and
usefulness of nitrous oxide,7 it is still commonly used in
anesthetic practice around the world. Because of its
limited potency, the usual practice is to administer
nitrous oxide at concentrations as high as 70% in
oxygen (inspired oxygen 30%) along with a potent
inhalational anesthetic agent (eg, sevoflurane) or intra-
venous propofol to produce a depth of anesthesia
sufficient for surgery. The prevailing view has been that
nitrous oxide is a cheap, relatively safe drug that can
reduce the exposure to other anesthetic drugs. We have
previously summarized the perceived risks and benefits
of nitrous oxide.7
Nitrous oxide interferes with vitamin B12 and folate
metabolism.8,9 It oxidizes the cobalt atom and irreversibly
inactivates the enzyme, methionine synthase. This
impairs production of methionine (from homocysteine),
used to form tetrahydrofolate and thymidine during DNA
synthesis. These adverse effects are time exposure
dependent and are probably greater in systemically
unwell patients.8 It is well established that prolonged
exposure to nitrous oxide can lead to a clinical syndrome
similar to pernicious anemia.7,10 Inhibition of methionine
synthase by nitrous oxide can be rapid and long-lasting:
Exposure beyond a few hours reduces methionine
synthase activity by 50%,8 and 12 to 24 hours of exposure
causes marked megaloblastic changes10 that can be
ameliorated by administration of sufficient folate or
vitamin B12.
Most relevant to those with coronary artery disease is
that nitrous oxide results in increased plasma homo-
cysteine and myocardial ischemia after surgery.11-13 In
addition, nitrous oxide can increase pulmonary artery
pressure and is contraindicated in pulmonary hyperten-
sion.7 Nitrous oxide activates the sympathetic nervous
system and can sensitize the myocardium to the
arrhythmogenic effects of epinephrine.14 Lastly, the
inclusion of nitrous oxide in the anesthetic gas mixture
increases the likelihood of intraoperative or postopera-
tive hypoxia.7
Nitrous oxide was the first anesthetic discovered and
remains a mainstay of anesthesia throughout the world;
it has probably been given to more than a billion
patients since 1844. It thus has a long history, and its
Myles et al 489
side effect and safety profile are thus thought to be well
understood. Its notable feature is that it allows a dose
reduction of other anesthetic drugs that are usually
more expensive and presumably more toxic. These
perceived benefits support critical evaluation of its
safety in a large clinical trial.15
Nitrous oxide, homocysteine, and
cardiovascular events
Nitrous oxideinduced inactivation of methionine
synthase increases plasma homocysteine after sur-
gery.11-13 Importantly, a recent study showed that plasma
levels of homocysteine remain elevated for at least a week
after surgery with nitrous oxide anesthesia.16 Long-term
elevation of plasma homocysteine concentration is an
independent risk factor for cardiovascular disease,17 and
an acute increase in plasma homocysteine causes
endothelial dysfunction18,19 and hypercoaguability.17
Consequently, hyperhomocysteinemia is associated with
increased risk of venous thromboembolism and stroke.20
Plasma homocysteine concentration is acutely raised
after oral methionine and is reliably associated with
endothelial dysfunction as measured by flow-mediated
vasodilation.18,19 Nitrous oxideinduced elevation of
homocysteine concentration may mirror that produced
by oral methionine, with a recent study demonstrating
nitrous oxide-based anesthesia significantly impaired
endothelial function as measured by flow-mediated
dilation in patients undergoing noncardiac surgery.13
Thus, nitrous oxide may also be a risk factor for
perioperative myocardial ischemia.
Badner et al11 randomly allocated 90 patients to
anesthesia with or without nitrous oxide. The nitrous
oxide group had significantly increased homocysteine
levels and a higher incidence of myocardial ischemia
(46% vs 25%, P b .05), more ischemic events (82 vs 53, P
b .02), and had more ischemic events lasting 30 minutes
(23 vs 14, P b .05). This is a particularly relevant finding
because it is known that the incidence of myocardial
ischemia is highest in the hours after surgery and that it is
strongly associated with postoperative MI.2
Our recent studies
We recently completed a moderate-sized multicenter
clinical trial of 2,050 patients undergoing noncardiac
surgerythe ENIGMA (Evaluation of Nitrous oxide In the
Gas Mixture for Anesthesia) Trial.15 Patients were
assigned to a nitrous oxide, or nitrous oxidefree,
anesthetic. We found that the avoidance of nitrous oxide
decreased the risk of wound infection (odds ratio [OR]
0.7, P = .034), severe vomiting (OR 0.4, P b .001), and
pneumonia (OR 0.5, P = .031). In addition, patients given
nitrous oxide had a greater length of stay in the intensive
care unit (ICU) (P = .02), suggesting an increased

490 Myles et al
incidence of more serious complications. These findings
were contrary to an earlier study that did not show these
differences.21 However, in the ENIGMA Trial, we did not
equalize the inspired oxygen concentrations in both
groups, and so, it is possible that some of the benefits of
avoiding nitrous oxide could be attributed to supple-
mental oxygen. The proposed trial (ENIGMA-II) will
address this issue.
The ENIGMA Trial was not designed for and had
insufficient power to detect a difference in the less
common, but serious complications of MI or death. We
nevertheless identified a possible increased risk of
confirmed MI in patients receiving nitrous oxide, 1.3%
versus 0.7% (adjusted P = .19). Interestingly, if patients
with new electrocardiogram (ECG) changes or cardiac
enzyme elevation, but not both (ie, unconfirmed MIs), are
included, there was a marked increase in the nitrous
oxide group, 30 versus 10 cases (P = .002); this requires
further study. There were 10 postoperative deaths (1.0%)
in the nitrous oxide group and 4 (0.4%) in the control
group (P = .26).
ENIGMA participants were an unselected group of
surgical patients, but the findings suggest that nitrous
oxide may be particularly detrimental in those at risk of
cardiac events.
In summary, there is an established association between
nitrous oxide and hyperhomocysteinemia, and flow-
mediated vasodilation, after surgery. There is also
evidence from a small clinical trial and post hoc results
from the ENIGMA Trial to suggest that avoidance of
nitrous oxide may lead to a reduction in postoperative
cardiac events and mortality.
Trial objectives
The overall study goal of the ENIGMA-II Trial is to
assess whether removal of nitrous oxide from the
anesthetic gas mixture reduces the incidence of major
perioperative cardiac events in patients with, or at risk
of, coronary artery disease who are undergoing major
noncardiac surgery. Specifically, we will test the
hypothesis that in inpatients undergoing anesthesia for
major noncardiac surgery, avoidance of nitrous oxide
will reduce the incidence of death and major
cardiovascular events. The study is funded by the
Australian National Health and Medical Research
Council (ID 436677).
Trial design
The ENIGMA-II study is a multicenter, international,
randomized, parallel-group, controlled trial, with patients
randomly allocated to either nitrous oxidecontaining
(70% nitrous oxide in oxygen [inspired oxygen fraction
0.3]) or nitrous oxidefree (70% nitrogen in oxygen
[inspired oxygen fraction 0.3]) anesthetic.
American Heart Journal
March 2009
Study methods
Patient population
ENIGMA-II will evaluate patients undergoing noncar-
diac surgery who are at risk of a perioperative major
cardiac event (Table I). We will enroll 7,000 patients in
30 to 40 participating sites in Australasia, Asia, North
America, and Europe. Centers will obtain institutional
review board approval before enrolling patients, and all
patients must provide informed consent to participate
in ENIGMA-II.
Assessment for eligibility
After first obtaining agreement from anesthesiologists
at each site, all elective noncardiac surgery patients are
screened for eligibility. The patients who are eligible but
not recruited into the trial are recorded in a study log that
includes the reasons for the lack of participation.
Allocation and randomization
After the patient's consent has been obtained, center
personnel will telephone a central 24-hour interactive
voice recognition system that prompts the researcher or
study coordinator to identify their study site, and
randomly allocates patients to a treatment group (1:1)
from a computer-generated list. Randomization is strati-
fied by site. The sample size is sufficiently large to ensure
comparable baseline characteristics, including surgical
risk factors, surgical technique, anesthetic technique, and
other aspects of perioperative care. ENIGMA-II is an
intention-to-treat trial. Any participant who is enrolled
and randomized to treatment group is followed for the
duration of the trial.
Definitions of end points
Primary end point
The primary end point is a composite of death and
major nonfatal cardiovascular events (MI, cardiac
arrest, pulmonary embolism, and stroke) at 30 days
after surgery.
Secondary end points
Secondary end points include the following: (i) MI, (ii)
cardiac arrest, (iii) pulmonary embolism, (iv) stroke, (v)
wound infection, (vi) severe nausea and vomiting, all up
to 30 days after surgery; plus (vii) duration of stay in the
ICU and hospital. For specific definitions, please see
Table II.
Data pertaining to study end points are sent to a
separate end point adjudication committee (see below).
Death: All deaths within 30 days of surgery from
any cause.
Myocardial infarction: requires any one of the
following criterion:
1. A typical rise of troponin or a typical fall of an
elevated troponin with at least one value above the
American Heart Journal
Volume 157, Number 3
Myles et al 491

Table I. Specific inclusion and exclusion criteria Table II. Baseline characteristics for the first 929 patients
recruited into the ENIGMA-II Trial (% unless otherwise specified)
Inclusion criteria
Variable
1. Adult males and females aged 45 years, undergoing noncardiac
surgery and general anesthesia expected to exceed 2 h Age, mean (SD), y 70 (9.4)
2. At increased risk of cardiac events, defined as any of the following: Male sex 66
a) History of coronary artery disease as indicated by a history of any ASA physical status
one of the following: I 0.6
i. Angina II 33
ii. MI II 62
iii. Segmental wall motion abnormality on echocardiography or a IV 3.9
fixed defect on radionuclide imaging Impaired exercise ability 28
iv. A positive exercise stress test for cardiac ischemia Preexisting major medical conditions
v. A positive radionuclide exercise, echocardiographic exercise, or Hypertension 84
pharmacological cardiovascular stress test for cardiac ischemia Coronary artery disease 38
vi. Coronary revascularization (CABG or PTCA) Previous MI 25
vii. Angiographic evidence of atherosclerotic stenosis 50% of the Previous CABG or angioplasty 21
diameter of any coronary artery Myocardial ischemia on perfusion scan 13
viii. ECG with pathological Q waves in 2 contiguous leads Heart failure 9.3
b) Heart failure Peripheral vascular disease 30
c) Cerebrovascular disease thought due to atherothrombotic disease Previous stroke or TIA 19
d) Undergoing aortic or peripheral vascular surgery; or High cholesterol 45
e) 3 or more of the following risk factors: Diabetes 36
Age 70 y Asthma/COPD 17
Any history of congestive heart failure Infection/Fever 3.9
Diabetes and currently taking an oral hypoglycemic agent or insulin Other 36
therapy Medications
Current treatment for hypertension Aspirin within the past 5 d 32
Preoperative serum creatinine N175 mol/L (N2.0 mg/dL) NSAID within the past 2 d 5.8
Current or previous high total cholesterol 6.2 mmol/L Clopidogrel 3.6
(N240 mg/dL) Ticlopidine 0.2
History of a transient ischemic attack (ie, a transient focal neurological Warfarin last 7 d 4.7
deficit that lasted b24 h and thought to be vascular in origin) Heparin 7.8
Emergency/Urgent surgery (ie, surgery which must be undertaken COX-II inhibitor 2.4
within 24 h of acute presentation to hospital) Nitrate 11
High-risk type of surgery (ie, intrathoracic or intraperitoneal) Statin 56
ACE inhibitor/ARB 54
Exclusion criteria Amiodarone 2.0
1. Having cardiac surgery -Blocker 41
2. Marked impairment of gas-exchange expected to require Diuretic 25
Fio2 N0.5 intraoperatively Calcium channel blocker 36
3. Specific circumstances where nitrous oxide is contraindicated Digoxin 3.9
(eg, volvulus, pulmonary hypertension, raised intracranial pressure) or Insulin 9.6
the anesthesiologist plans to use supplemental oxygen (eg, colorectal Oral hypoglycemic 26
surgery) Antibiotic 7.4
4. Nitrous oxide is unavailable for use
ASA, American Society of Anesthesiologists; TIA, transient ischemic attack; COPD,
PTCA, Percutaneous transluminal coronary intervention. chronic obstructive pulmonary disease; NSAID, nonsteroidal anti-inflammatory drug;
COX, cyclooxygenase; ACE, angiotensin-converting enzyme; ARB, angiotensin
receptor blocker.
Estimated b3 metabolic equivalents.

99th percentile of the upper reference limit, detected
at its peak post surgery in a patient without a
documented alternative explanation for an elevated
troponin (eg, pulmonary embolism).22 This criterion
also requires that one of the following must also exist:
A. ischemic signs or symptoms (ie, chest, arm, neck,
or jaw discomfort; shortness of breath, pulmon-
ary edema)
B. development of pathologic Q waves present in
any 2 contiguous leads that are 30 milliseconds
C. the ECG changes indicative of ischemia (ie, ST
segment elevation [2 mm in leads V1, V2, or V3
or 1 mm in the other leads], ST segment
depression [1 mm], or symmetric inversion of
T waves 1 mm) in at least 2 contiguous leads
D. coronary artery intervention (ie, percutaneous
coronary intervention or coronary artery bypass
graft surgery [CABG] surgery)
E. new or presumed new cardiac wall motion
abnormality on echocardiography or new or pre-
sumed new fixed defect on radionuclide imaging
2. Pathologic findings of an acute or healing MI
3. Development of new pathological Q waves on an
ECG if troponin levels were not obtained or were
obtained at times that could have missed the
clinical event

492 Myles et al
Cardiac arrest: Defined as a successful resuscitation
from either documented or presumed ventricular
fibrillation or sustained ventricular tachycardia
or asystole.
Pulmonary embolism: high probability on VQ scan or
documented on pulmonary angiogram or spiral com-
puted tomography or at autopsy.
Stroke: cerebral infarction or intracerebral hemorrhage
on computed tomography or magnetic resonance ima-
ging scan, or new neurological signs (paralysis, weakness,
or speech difficulties) lasting N24 hours or leading to
earlier death.
Severe nausea and vomiting: at least 2 episodes of
severe nausea or vomiting N6 hours apart, or if requiring
N2 doses of antiemetic medication.
Wound infection: associated with purulent discharge
and/or a positive microbial culture.
ICU stay: including initial ICU admission and additional
time after any readmission.
Hospital stay: from the start (date, time) of surgery
until actual hospital discharge.
End point adjudication committee
The committee consists of an internal medicine
physician with board-equivalent anesthesiology training
and 2 cardiac anesthesiologists. Their role is to verify each
of the study end points and to resolve any uncertainty as
to any of the above outcomes.
Surgical and anesthetic techniques
Preoperative demographic characteristics and
details of each patient's medical and surgical history are
recorded. They will also undergo a 12-lead ECG,
chest x-ray, pathology testing, and other routine
investigations.
On the day of surgery, patients are allocated to 1
of the 2 treatment groups. All other perioperative
clinical care is according to standard practice at each
site, as this is an effectiveness trial designed to
represent real-world practice.23 This includes choice
of anesthetic drugs, analgesic regimens and/or regio-
nal analgesia techniques, antiemetics, and initiation or
continuation of perioperative cardiac medications
including anticoagulant and antiplatelet therapy. These
vary and are allowable in the trial. All such relevant
perioperative data are recorded on the study case
report form.
Blood is collected postoperatively at 6 to 12, 24, 48,
and 72 hours for cardiac enzyme (troponin and/or
creatine kinase-MB) levels, and a 12-lead ECG is
performed on days 1 and 3 after surgery to detect MI.
Additional tests are ordered if clinically indicated (eg,
chest pain, dyspnea, circulatory instability). In addition,
patients are contacted by telephone at 30 days and
their medical record reviewed to ascertain if they have
experienced any adverse outcomes.
American Heart Journal
March 2009
Study medications and duration
After induction of anesthesia, patients allocated to the
nitrous oxide group will have their anesthesia main-
tained with 70% nitrous oxide (inspired) in 30% oxygen
and 1 of 4 other hypnotic agents (isoflurane, sevoflur-
ane, desflurane, or propofol) at the discretion of the
anesthesiologist. Patients allocated to the nitrous oxide
free group will receive 30% oxygen in an oxygen/
medical air mixture and have their anesthesia main-
tained with one of the above hypnotic agents (also
according to anesthesiologist preference). In both
cases, depth of anesthesia can be adjusted according to
clinical judgment and/or data from depth of anesthesia
monitoring devices (if available). All other perioperative
clinical care will be according to standard practice. All
relevant factors will be recorded on a trial case
report form.
Study procedures, blinding,
and follow-up
Anesthesiologists will have knowledge of group after
randomization (for safe use of nitrous oxide), but
administration and group identity will be concealed
from the surgeon. Anesthesiologists participating in the
conduct of the study will receive education explaining
the importance of, reasoning for, and methods to
ensure blinding from surgeons and others. The
anesthetic machine flow meters will be concealed
from the surgical staff, using drapes or cardboard
screen. To ensure blinding of all the other staff, the
anesthesiologist will place the original anesthesia
record after surgery in a sealed opaque envelope; the
envelope is to be stored in the patient's hospital
record and should not be opened until after the 30-
day follow-up unless there is a clinical imperative.
Patients, surgeons, and research staff collecting data
and interviewing patients postoperatively will be blind
to treatment allocation.
Collection of data
Data are collected by local research staff and
entered onto a paper case report form. All data are
subsequently entered onto a database on the study
Web site (www.enigma2.org.au). This is managed by
Monash University's Center for Clinical Research
Excellence in Therapeutics (Melbourne, Australia),
where all data and processes are reviewed each day
at the data management center (see below). Data
fields are checked, and in conjunction with the local
site research staff, missing data or inconsistencies are
corrected, before being automatically downloaded on
to a confirmed database. At the end of the trial, site-
specific data will be sent to each site investigator on
a CD-ROM for long-term storage.
American Heart Journal
Volume 157, Number 3
Myles et al 493

Statistical considerations Table III. Types of surgery, anesthetic techniques, and

intraoperative nonanesthetic drug administration for the first 929
Sample size and power patients recruited into the ENIGMA-II Trial (%)
The recently revised international guidelines for the
Type of surgery
diagnosis of MI22 should result in an increase in cardiac Vascular 32
events because of increased sensitivity (ie, identifying Orthopedic 16
small infarcts). The sample size is based on a clinically Colorectal 15
important (25%) reduction in cardiac events or death, General (noncolorectal) 13
Neurosurgery
from 8% to 6%. These estimates are based on our
Spinal 8.1
previous research and large observational studies.5,24 A Craniotomy 1.5
type I error of 0.05 and a type II error of 0.1 require Urology 8.8
7,000 patients to be included in this study. We tested Head and neck 5.2
these assumptions with our ENIGMA Trial database. Plastics 1.2
Intraoperative cardiac drugs
When the above trial entry criteria were used, we
-Blocker 33
identified 656 comparable patients (313 nitrous oxide Clonidine 3.3
group, 343 control). Nitrous oxide was associated with Antiemetic prophylaxis 62
a higher incidence of reported MI (OR 2.4, 95% Dexamethasone 58
confidence interval [CI] 1.1-5.3, exact P = .029) and a 5-HT3 antagonist 4.7
Droperidol/Haloperidol 16
trend to increased mortality (OR 7.3, 95% CI 0.80-50, Other
exact P = .07). Major regional block 26
Bispectral index monitoring 51
Statistical methods
All patients who are randomly allocated to study drug the true incidence of cardiac events and death in this
administration will be considered as comprising the population and will provide important information for
intention-to-treat population for all primary, secondary, future patients and their families, as well as for govern-
and safety analyses. Baseline characteristics of the 2 ment and other health care agencies. The relative clinical
treatment groups will be tabulated using appropriate importance can be determined (impact on daily living,
summary statistics. social activities, requirement for rehospitalization, long-
Stratification will be by center, but not other factors as term disability). In addition, the most important risk
these are likely to be balanced in this large trial. Intention- factors for serious postoperative complications can be
to-treat analyses will be undertaken. The cumulative identified. Thus, the database will be examined using a
incidence of the combined end point will be analyzed variety of exploratory regression techniques to address
using Fisher exact test, with covariate adjustment each of these issues.
performed by logistic regression. Results will be We also plan to conduct an exploratory analysis for
expressed with risk ratios and exact 95% CIs. Other cointerventions believed to reduce perioperative cardiac
secondary end points will be compared with Cox events. There are several interventions believed to reduce
proportional hazards and Wilcoxon rank sum tests. perioperative cardiac events in surgical patients with
Two interim analyses are planned for the ENIGMA-II known or suspected coronary artery disease,3,5 but all
Trial. These will be performed after enrolment of 3,000 have been based on small trials or meta-analysis of small
and 5,000 patients, that is, at 43% and 71% of the target trials. We plan to use propensity scores and logistic
recruitment number of 7,000. The study statistician (A. F.) regression analysis to investigate the following factors:
will perform the interim analyses, and the results will perioperative -blockade,5,6 perioperative statin thera-
discussed by the Data Safety and Monitoring Committee py,25 intraoperative hypothermia,26 spinal or epidural
(DSMC), according to an agreed charter. The interim local analgesic block,27,28 and perioperative calcium
analysis will be adjusted according to an O'Brien and antagonists,29 with corresponding null hypotheses that
Fleming Type I error spending function, using a none of these factors are associated with the incidence of
combined P value b.05. postoperative cardiac events in surgical patients with
elevated cardiovascular risk.
Planned substudies
ENIGMA-II will support a cohort study of the incidence,
clinical importance, risk factors, and economic costs of Data safety and monitoring committee
cardiac events and mortality in patients undergoing The DSMC consists of a clinical epidemiologist (Chair),
anesthesia for major surgery. This will provide a unique anesthesiologist independent statistician, and a clinical
opportunity to collect extensive and accurate data on pharmacologist. The DSMC will discuss the interim
7,000 surgical patients with known or suspected cor- results and vote for continuation or stopping the trial.
onary artery disease. This will allow reliable estimation of This will be communicated to the Steering Committee

494 Myles et al
according to a stopping rule of P b .001 for the primary
study end point and consideration of other evidence
relevant to the recommendation of the DSMC. The
conduct of the DSMC is to be guided by the paper by
DeMets et al.30
Current status of the trial
ENIGMA-II is currently recruiting patients in 23 centers
within 5 countries and has randomized 1,258 patients as
of November 2008. Tables II and III report on character-
istics of the first 929 patients enrolled in the trial. These
data demonstrate a cohort of patients at increased risk of
cardiovascular events after surgery.
Conclusion
Exposure to nitrous oxide impairs methionine
synthase, folate and DNA production, and increases
homocysteine levels. These adverse effects are likely to
be enhanced in at-risk patients; we will thus focus our
trial on patients with coronary artery disease and those
with risk factors for coronary artery disease. When
considering the widespread use of nitrous oxide around
the world, small differences in outcome would have
major implications for surgical practice.
Disclosures
None of the authors has declared any conflict of
interest.
References
1. Weiser TG, Regenbogen SE, Thompson KD, et al. An estimation of the
global volume of surgery: a modelling strategy based on available
data. Lancet 2008;372:139-44.
2. Mangano DT. Perioperative cardiac morbidity. Anesthesiology 1990;
72:153-84.
3. Devereaux PJ, Goldman L, Cook DJ, et al. Perioperative cardiac events
in patients undergoing noncardiac surgery: a review of the magnitude
of the problem, the pathophysiology of the events and methods to
estimate and communicate risk. CMAJ 2005;173:627-34.
4. Fleisher LA, Corbett W, Berry C, et al. Cost-effectiveness of differing
perioperative beta-blockade strategies in vascular surgery patients.
J Cardiothorac Vasc Anesth 2004;18:7-13.
5. Devereaux PJ, Yang H, Yusuf S, et al. Effects of extended-release
metoprolol succinate in patients undergoing non-cardiac surgery
(POISE trial): a randomised controlled trial. Lancet 2008;371:1839-47.
6. Mangano DT, Goldman L. Preoperative assessment of patients with
known or suspected coronary disease. N Engl J Med 1995;333:
1750-6.
7. Myles PS, Leslie K, Silbert B, et al. A review of the risks and benefits of
nitrous oxide in current anaesthetic practice. Anaesth Intensive Care
2004;32:165-72.
8. Nunn JF. Clinical aspects of the interaction between nitrous oxide and
vitamin B12. Br J Anaesth 1987;59:3-13.
9. Maze M, Fujinaga M. Recent advances in understanding the actions
and toxicity of nitrous oxide. Anaesthesia 2000;55:311-4.
American Heart Journal
March 2009
10. Amos RJ, Amess JA, Hinds CJ, et al. Incidence and pathogenesis of
acute megaloblastic bone marrow change in patients receiving
intensive care. Lancet 1982;2:835-8.
11. Badner NH, Beattie WS, Freeman D, et al. Nitrous oxide-induced
elevated homocysteine concentrations are associated with increased
myocardial ischemia in patients undergoing carotid endarterectomy.
Anesth Analg 2000;91:1073-9.
12. Myles PS, Chan MTV, Leslie K, et al. Effect of nitrous oxide on plasma
homocysteine and folate in patients undergoing major surgery. Br J
Anaesth 2008;100:780-6.
13. Myles PS, Chan MTV, Kaye DM, et al. Effect of nitrous oxide
anesthesia on plasma homocysteine and endothelial function.
Anesthesiology 2008;109:657-63.
14. Hynes MD. Catecholamine mechanisms in the stimulation of mouse
locomotor activity by nitrous oxide and morphine. Eur J Pharmacol
1983;90:109-14.
15. Myles PS, Leslie K, Chan MTV, et al. Avoidance of nitrous oxide for
patients undergoing major surgery: a randomized controlled trial.
Anesthesiology 2007;107:221-31.
16. Ermens AA, Refsum H, Rupreht J, et al. Monitoring cobalamin
inactivation during nitrous oxide anesthesia by determination of
homocysteine and folate in plasma and urine. Clin Pharmacol Ther
1991;49:385-93.
17. Mayer EL, Jacobsen DW, Robinson K. Homocysteine and coronary
atherosclerosis. J Am Coll Cardiol 1996;27:517-27.
18. Bellamy MF, McDowell IF, Ramsey MW, et al. Hyperhomocysteine-
mia after an oral methionine load acutely impairs endothelial function
in healthy adults. Circulation 1998;98:1848-52.
19. Chambers JC, McGregor A, Jean-Marie J, et al. Demonstration of rapid
onset vascular endothelial dysfunction after hyperhomocysteinemia: an
effect reversible with vitamin C therapy. Circulation 1999;99:1156-60.
20. Wald DS, Law M, Morris JK. Homocysteine and cardiovascular
disease: evidence on causality from a meta-analysis. BMJ 2002;325:
1202-9.
21. Fleischmann E, Lenhardt R, Kurz A, et al. Nitrous oxide and risk of
surgical wound infection: a randomised trial. Lancet 2005;366:1101-7.
22. Thygesen K, Alpert JS, White HD, et al. Universal definition of
myocardial infarction. Circulation 2007;116:2634-53.
23. Tunis SR, Stryer DB, Clancy CM. Practical clinical trials: increasing the
value of clinical research for decision making in clinical and health
policy. JAMA 2003;290:1624-32.
24. Raby KE, Goldman L, Creager MA, et al. Correlation between
preoperative ischemia and major cardiac events after peripheral
vascular surgery. N Engl J Med 1989;321:1296-300.
25. Kapoor AS, Kanji H, Buckingham J, et al. Strength of evidence for
perioperative use of statins to reduce cardiovascular risk: systematic
review of controlled studies. BMJ 2006;333:1149-56.
26. Frank SM, Fleisher LA, Breslow MJ, et al. Perioperative maintenance
of normothermia reduces the incidence of morbid cardiac events.
JAMA 1997;277:1127-34.
27. Rodgers A, Walker N, Schug S, et al. Reduction of post-operative
mortality and morbidity with epidural or spinal anaesthesia: results
from overview of randomised trials. BMJ 2000;321:1493-7.
28. Beattie WS, Badner NH, Choi PT. Meta-analysis demonstrates
statistically significant reduction in postoperative myocardial infarc-
tion with the use of thoracic epidural analgesia. Anesth Analg 2003;
97:919-20.
29. Wijeysundera DN, Beattie WS, Rao V, et al. Calcium antagonists are
associated with reduced mortality after cardiac surgery: a propensity
analysis. J Thorac Cardiovasc Surg 2004;127:755-62.
30. DeMets DL, Pocock SJ, Julian DG. The agonising negative trend in
monitoring of clinical trials. Lancet 1999;354:1983-8.

494.e1 Myles et al
Appendix A. ENIGMA-II trial
organization and Committees
Sponsor: Alfred Hospital, Melbourne, Australia
Funding sources: Australian National Health and
Medical Research Council (NHMRC) project grant
ID 436677
Steering Committee: Paul Myles (Chair and Principal
Investigator), Kate Leslie, Phil Peyton, Mike Peach,
Brendan Silbert, Philip J Devereaux, Dan Sessler, Andrew
Forbes, Scott Beattie; Research Manager, Sophia Wallace
Chief and Associate Investigators: Paul Myles, Kate
Leslie, Phil Peyton, Mike Peach, Brendan Silbert,
Philip J. Devereaux, Dan Sessler, Andrew Forbes
American Heart Journal
March 2009
End Point Adjudication Committee: James W.
Tomlinson, David R. McIlroy, Neal Badner
Clinical Pharmacologist: Henry Krum
Statistician: Andrew Forbes
Data and Safety Monitoring Committee: Konrad
Jamrozik (Chair), John Rigg, Henry Krum; Independent
Statistician, Philip Ryan
Data Management and Quality Control: Sophia Wallace,
Adam Meehan
Website Design and Maintenance: Adam Meehan