Clinical Epidemiology and Health Service Evaluation Unit

Evidence Report

Excluding acute cardiac ischaemic

events in patients presenting to
emergency departments with chest
pain

Adrian Lowe
Associate Professor Donald Campbell
Jill Nosworthy

Clinical Epidemiology and Health Service Evaluation Unit

Ground Floor, Connibere Building
Royal Melbourne Hospital
Grattan Street, Parkville 3050
Telephone: +61 3 9342 8772
Facsimile: +61 3 9342 7060

Table of Contents
LIST OF ABBREVIATIONS .....................................................................................ii
SUMMARY ..................................................................................................................1
Task............................................................................................................................1
Conclusions and recommendations............................................................................1
Feasibility...................................................................................................................1
METHODOLOGY ......................................................................................................1
Search Strategy ..........................................................................................................1
Resources Searched ...................................................................................................2
Refinements, Searching & Reporting Constraints .....................................................2
CLINICAL PRACTICE GUIDELINES....................................................................3
HEALTH TECHNOLOGY ASSESSMENT.............................................................4
Published Primary Literature .....................................................................................4
MEDLINE Search Strategy Used ..........................................................................4
Primary Publications Summary .................................................................................5
Non-standard Lead ECG........................................................................................5
Exercise Stress Testing ..........................................................................................5
Biomarkers.............................................................................................................5
Single sample measures .........................................................................................6
Serial Measurements..............................................................................................6
Technetium-99m Sestamibi Myocardial Perfusion Imaging .................................6
Electron beam computed tomography ...................................................................7
Goldman Chest Pain Protocol................................................................................7
Other Computer-Based Decision Aids...................................................................7
Left bundle-branch blocks (LBBB) sub-population specific studies.....................7
ORIGINAL ARTICLES REVIEWED ......................................................................8
Appendix 1: Summary of guidelines concerning AMI............................................35
Appendix 2: Summary of AHRQ Evidence Report No. 26.....................................57
Appendix 3 Abstracts of all primary publications which were critically appraised 63
Appendix 4 Disclaimer and NHMRC Levels of Evidence .....................................78

LIST OF ABBREVIATIONS
AHRQ

Agency for Healthcare Research and Quality

ACEP

American College of Emergency Physicians

SRPC

Society of Rural Physicians of Canada

ED

Emergency Department

LOS

Length of stay

WHO

World Health Organisation

AMI or MI

Acute Myocardial Infarction or Myocardial Infarction

ACI

Acute Cardiac Ischaemia

cTnI

Cardiac Troponin I

cTnT

Cardiac Troponin T

CK

Creatine Kinase

CK-MB

Creatine kinase MB isoenzyme

ESE

Exercise stress echocardiography

ECG

Electrocardiogram

SECG

Serial electrocardiograms

LBBB

Left bundle branch block

EBCT

Electron-beam computed tomography

99mTc-GLA

Technetium-99m labelled D - Gluaric Acid

ii

SUMMARY
Task
The Clinical Epidemiology & Health Service Evaluation Unit was asked to review the
research literature on the best practice in excluding an acute ischemic event when
patients present to an emergency department with chest pain.

Conclusions and recommendations

This literature review identified 4 guidelines relevant to this topic. A summary of
each of these guidelines can be found in Appendix 1. The latest guideline published
in 2000 by the American College of Emergency Physicians (ACEP) provides a very
good summary of evidence concerning diagnosis of AMI and practical diagnostic
regimens for use in patients presenting to ED with chest pain.
In 1998, ACEP produced a document outlining all possible causes of chest pain. ED
staff are likely to find this document most useful as a reference for making a
differential diagnosis when faced with ambiguous chest pain symptoms. The other
two guidelines are of interest but are not as up to date as the ACEP 2000 clinical
policy incorporating guidelines for patient management.
The AHRQ health technology assessment (AHRQ 2001) considered the accuracy of
technologies for diagnosing chest pain in the ED. It reviewed their clinical impact
when used in this setting and contains rigorous meta-analytic overviews of both the
accuracy of the technology and its clinical impact. The information from this review
is directly relevant to the management of chest pain in ED (see Appendix 2).
A critical appraisal of the primary publications from 1999 - 2001 in this area did not
add a great deal to that already identified, especially by the AHRQ review (see
Appendix 3). However it did add further weight to the recommendations from the
AHRQ about the use of serial biomarker measurement providing an opportunity for
the development of effective timed myocardial infarction rule-out protocols.

Feasibility
Suitable material for informing a best practice approach to excluding an acute
ischemic event when patients present to an emergency department with chest pain was
presented in the AHRQ 2001 report and ACEP 2000 clinical policy.

METHODOLOGY
Search Strategy
The Clinical Epidemiology & Health Service Evaluation Unit defines best available
evidence as that research we can identify that is least susceptible to bias. This was
determined through applying the predefined NHMRC criteria (see Appendix 4).
The search commenced with systematic reviews, evidence-based clinical guidelines or
health technology assessments and randomised controlled trials. If sound relevant
material of this type is identified then the search ceased. Otherwise the search

strategy is broadened to include studies which are more prone to bias making them
less generalisable. Case-control and longitudinal cohort studies are included in our
critical appraisal reports. Other studies, which have less methodical rigour, are
included but not critically appraised.

Resources Searched
We searched the following databases and Internet websites:

Cochrane Library CD-ROM

Medline (OVID)

CINAHL (OVID)

SumSearch http://sumsearch.uthscsa.edu/searchform4.htm

National Guidelines Clearinghouse (http://www.guidelines.gov), and other

evidence-based sites known to store guidelines

NHS Centre for Reviews and Dissemination (NHS CRD)

http://www.nelh.nhs.uk/guidelines_database.asp.

Agency for Healthcare Research and Quality (http://www.ahrq.gov)

Refinements, Searching & Reporting Constraints

We only included articles published since 1999, and applied the following inclusion
and exclusion criteria:
Inclusion Criteria

Focus on all patients with chest pain in the emergency department;

Published primary studies;

Published clinical practice guidelines (whether generated through evidencebased methods or through consensus)
Exclusion Criteria

Study examined less than five patients;

Study was published in a language other than English;

Study presented data included in another published report;

Study examined long term therapy not possible to implement in ED.

CLINICAL PRACTICE GUIDELINES

A total of four guidelines were identified that addressed treatment of all chest pain.
The descriptive characteristics of these guidelines are shown in Table 1. Where
available, we include a brief summary of cited guidelines in Appendix 1.
Table 1. Description of guidelines cited
Characteristic
Developers
Title
Outcomes Considered for ED
Methods to Collect Evidence
Methods to Analyse Evidence
Length
URL
Comments

ACEP, 2000
American College of Emergency of Physicians
Clinical Policy: Clinical issues in the evaluation and management of adult patients
presenting with suspected acute myocardial infarction or unstable angina.
ECG eligibility, role of angioplasty, use of serum markers, serial 12 lead ECG,
chest pain evaluation units.
MEDLINE search between 1993-1998
Review by expert subcommittee
26 pages
http://www.acep.org/library/files/pdf/cp402120.pdf
Excellent summary of evidence concerning diagnosis of AMI, and practical
diagnostic regime.

Characteristic
Developers
Title
Outcomes Considered for ED
Methods to Collect Evidence
Methods to Analyse Evidence
Length
URL
Comments

ACEP, 1998
American College of Emergency Physicians
Chest Pain: Clinical Decision-Making
Not stated
Not stated
Not stated
13 pages
http://cghane.netfirms.com/chest%20pain.pdf
Limited referencing, but good source of information for all possible conditions
that may cause chest pain

Characteristic
Developers
Title

ACEP, 1995
American College of Emergency Physicians
Clinical Policy for the initial approach to adults presenting with a chief complaint
of chest pain, with no history of trauma
Adult population, presenting with chest pain. Does not include patients, with
history of proximate trauma, or visible lesions in chest area, individuals with very
brief periods of pain
Correct diagnosis of chest pain
Not stated
Expert panel
26 pages
http://www.acep.org/library/files/pdf/cp402050.pdf or Ann Emerg Med, 1995;
25: 274-299
This document is a number of years old now, but has a very useful quick
reference form attached for diagnosis and treatment of chest pain. Includes
conditions other than AMI.

Target Population
Outcomes Considered for ED
Methods to Collect Evidence
Methods to Analyse Evidence
Length
URL
Comments

Characteristic
Developers
Title
Target Population
Outcomes Considered for ED
Methods to Collect Evidence
Methods to Analyse Evidence
Length
URL
Comments

SRPC, 1999
Society of Rural Physicians of Canada
Chest pain guideline and continuous quality improvement system for Canadian
rural emergency health care facilities
Adult population, presenting with chest pain to a rural hospital setting.
Rapid identification and initial management of patients with possible acute
coronary ischemic syndromes and acute myocardial infarction.
MEDLINE searches
Expert panel
1 page flow chart with an 11 page supporting document
http://www.cma.ca/cjrm/vol-4/issue-1/0009.htm
Or Canadian Journal of Rural Medicine, 1999; 4(1) 9-19Ann
Only recommends ECG and clinical symptoms for diagnosis of AMI or ACI.

HEALTH TECHNOLOGY ASSESSMENT

A comprehensive health technology assessment report published in 2001, by the
Agency for Healthcare Research and Quality (AHRQ) was identified. This report
considered the accuracy of technologies for diagnosing chest pain in the ED and their
clinical impact when used in this setting. In addition, the report also conducted
decision and cost-effectiveness (CE) analyses to investigate the interactions between
technologies diagnostic performances and costs, populations and outcomes. An
extensive summary of the report is presented in Appendix 2.
Characteristic

AHRQ, 2001

Developers

Agency for Healthcare Research and Quality

Title

Evaluation of Technologies for Identifying Acute Cardiac Ischemia in Emergency

Departments. Evidence Report/Technology Assessment Number 26. AHRQ
Publication No. 01-E006, May 2001

Target Population

Studies examining the diagnostic value of technologies in assessing risk of ACI,

AMI, and UAP, in the Emergency Department

Outcomes Considered for ED

Unnecessary hospitalisations; test sensitivity and specificity.

Methods to Collect Evidence

MEDLINE search for articles published between 1966 and December 1998.

Methods to Analyse Evidence

Meta analysis and systematic review

Length

315 pages

URL

http://www.ahrq.gov/clinic/aciscinv.htm

Comments

Excellent summary of the technologies used to diagnose AMI and ACI, and the
evidence support for them.

Published Primary Literature

An update of evidence was conducted for new original articles, using a modified
version of the search strategy employed by Ioannidis, et al. (AHRQ, 2001), in the
MEDLINE database.
MEDLINE Search Strategy Used
Setting (any of)
Emergency medical service communication systems/or Emergency medical
services/or Transportation of patients/
Exp mobile health units/or prehospital.mp.
Exp telecommunications/
Exp transportation of patients/
Exp emergency medical technicians/
Exp Emergency service, hospital/
Emergenc$.tw.
Disease (any of)
Exp myocardial ischemia/di [Diagnosis]
Exp chest pain/di [Diagnosis]
(chest pain or myocardial infarction or myocardial ischem$).tw
Technology
Exp sensitivity and specificity/

Population
Limit to human
Limits (all of)
Yr 1999 2001
Limit to English language
Not review
Not letter
Not editorial
Outcome of the Search
The search lead to 101 documents being identified. Abstracts were reviewed for all
and 42 documents met the inclusion criteria and full articles were sought. Of these,
one was not in English, four were review articles (despite the search strategy used)
and one could not be located prior to the review being completed. Therefore 36
articles were reviewed. Critical appraisal of the individual articles is located in the
following pages. A composite summary of the findings is included below and
citations and abstracts can be found in Appendix 3.
The conclusion from the critical appraisal of these articles is that the evidence for the
use of serial measurements of biomarkers in ED rather than single sample
measurement is becoming stronger. It is pointing towards the development of an
accurate and efficacious rule out protocol for myocardial infarction within 6 hours
of presentation at ED (see Herren et al, 2001). However more work is required over a
longer time period before the recommendations made by Herren and colleagues could
be adopted.

Primary Publications Summary

Non-standard Lead ECG
Practically all of the cited articles have used ECG measurements to some degree.
However, only Aufderheide, et al (2000) have recently focused on a part of ECG as
the primary focus of analysis. This well conducted prospective cohort study supports
the use of computerised detection of QT- dispersion and ST-segment deviations,
instead of manual interpretation of ECG findings.
Exercise Stress Testing
Five articles were cited for exercise stress testing. As a group, the quality was not
high. As all studies selected only very low risk groups, the prevalence of coronary
events was very low. Unfortunately, sample sizes were not high enough to record
sufficient coronary events for any confidence to make conclusions about the
effectiveness of this technology. All studies showed no adverse effects associated
with this technique. However, it is yet to be shown that this technique provides any
additional diagnostic value for very low risk patients. These results further support
the AHRQ (2001) findings.
Biomarkers
Considerable research has been conducted into the utility of biomarkers in predicting
AMI and ACI. As most of these papers examine a number of markers
5

simultaneously, these papers have been divided into single sample, or serial
measurements, and not test specific groups.
Single sample measures
To find a biomarker that can identify AMI or ACI at admission is an ideal. Three
papers cited have explored the utility of various biomarkers in this role. The
biomarkers examined were cTnI and cTnT, CK, and CK-MB, Myoglobin. The
quality of these studies was generally good.
The sensitivity of individual markers was poor (highest value reported was for cTnI at
53% sensitivity, with 91% specificity, Porela et al, 2000) indicating inadequate
predictive value for a rule out protocol. However, when combined with ECG and a
combination of markers, this value was shown to jump considerably. For example,
Porela, et al, (2000) found a combination of ECG, CK-MB or cTnI could give a
sensitivity of 90%, but with only a specificity of 54%. As such, using such a test could
identify 90% of all AMI cases, 46% of positive cases being incorrectly identified.
The use of combined measurements at initial evaluation is a relatively new
development, with only one study cited by AHRQ (2001) doing so.
Serial Measurements
There has been a substantial volume of work produced on the serial measurement of
markers, with the aim of improving test performance. The quality of this work is
heterogeneous, as are the results.
Some of the heterogeneity can be explained by the time frame over which the
repeated tests were taken. The work by Zimmerman, et al (1999) helps explain this,
as it characterises the variation in sensitivity and specificity of all the main biomarkers
over a 22-hour period following onset of chest pain. This work shows that CK-MB
subforms, myoglobin, cTnT, cTnI all have a maximal sensitivity at around 10 hours
after onset of pain. CK-MB and CK-MB mass increase sensitivity with time,
plateauing at around 10 hours.
Other studies have shown sensitivities of 91 and 94% (specificity 91.4 and 94%
respectively) within 2 hours, for measurement of change in CK-MB or change in CKMB and cTnI (Fesmire, 2000a, Fesmire, 2000b). These are very promising results, as
they suggest a very prompt rule out decision.
There are also other promising rule out protocols that take longer to reach a
decision. This is an important area of development since the studies reviewed by
AHRQ (2001).
Technetium-99m Sestamibi Myocardial Perfusion Imaging
Four new studies examined myocardial perfusion imaging as a diagnostic tool, but
only two could be used to report the sensitivity of this test. These well-conducted
studies showed very good performance (sensitivity of 92, and 100%, with specificities
of 67 and 86% respectively, Kontos, et al, 1999 & Paventi, 2001). These reported
specificities would still lead to a high number of patients being admitted
unnecessarily. These results are consistent with those of the AHRQ (2001) document.

Electron beam computed tomography

One study reported results from this new potential diagnostic tool for AMI (Laudon,
et al, 1999). This small study showed excellent sensitivity (100%) for AMI, there was
a relatively poor specificity (63%). This technique may need more development.
There were no studies reported in the AHRQ (2001) results that used this technique.
Goldman Chest Pain Protocol
The two studies reporting on the use of the Goldman protocol did not provide strong
support for the diagnostic tool, with information being very difficult to extract. In the
one study where sensitivity could be calculated, a value of 76.7% for AMI was
achieved with a specificity of 52.3%. The AHRQ (2001) cites stronger evidence for
this protocol.
Other Computer-Based Decision Aids
Aase (1999) reported on a new computer aided decision protocol. This program had
good results, with 84% sensitivity and 75% specificity for diagnosing AMI.
Left bundle-branch blocks (LBBB) sub-population specific studies
Two studies specifically studied a sub-population who presented to the ED with
LBBB. These studies both found that the ECG results had very low sensitivities in this
group for detecting AMI. ECG combined with CK-MB measurement did increase the
sensitivity of the diagnosis, but only to 63%. As such, patients with LBBB should be
treated with increased caution.

ORIGINAL ARTICLES REVIEWED

Non-standard Lead ECG
Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Aufderheide, et
al, 2000

Computerised 12lead ECGs with

QT-dispersion and
ST-segment
deviations
combined with
physician consensus

Country: USA
Median age:62
Age range: Not
stated
Enrolled: 1568
Evaluation: 1161
male (%): 54.7
Race: 79.6% White

Age >18
Stable condition.
Chest pain

Trauma

WHO criteria;
appropriate 12-lead ECG
changes CK-MB,
Troponin T or I results
based on established
criteria in hospital labs.

AMI 24.6

Specificity (%)

Study conclusions

Potential
verification bias

Limitations/comments

AMI 97
ACI 80

This study supports a

potential clinical role for
automated QT dispersion
when combined with other
diagnostic methods for
detecting AMI and ACI.
Physician consensus alone
has higher specificity
(99%), but lower sensitivity
(48%)

Not stated

Test criteria for Sensitivity (%)

AMI
Computerised
classification based
on achieving >85%
specificity for
detecting ACI and
AMI. QT end
threshold value was
>45ms. QT-peak
threshold was
>55ms

Original Articles Summaries

AMI 65
ACI 62

Exercise Stress Testing

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Buchsbaum, et
al, 2001

Exercise stress
echocardiography at
approximately 6 hrs

Country: USA
Mean age: Not
stated
Age range: Not
stated
Enrolled: 149
Evaluation: 148
male (%):Not stated
Race: Not stated

Low-risk patients:
Age 30 years
Normal CK-MB isoenzyme
Normal or non-diagnositic
ECG
No history of cardiac
disease

Not stated

30 and 60-day telephone

follow up to determine if
MI, cardiac death, or
cardiac revascularization
had occurred. Medical
records used to confirm
events.

Cardiac event: 2.7

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

New wall motion

abnormalities
observed on postexercise images
(1mm of ST
depression observed
at 80ms from J
point)

75

97.1

Exercise stress
echocardiography (ESE) can
be used to evaluate low-risk
chest pain patients in the
ED. Patients with a normal
ESE may be considered for
discharge with minimal risk
of sequelae.

Only four adverse events

observed. Small number
of events will result in
unstable estimates of
sensitivity.

Small number of events

observed. The sample was
selected at the discretion of
an attending physician, with
poor description of criteria
used.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Geleijnse, et al,
2000

Early dobutamineatropine stress

echocardiography if
CK and ECG
results were
negative

Country: Europe
Mean age:58
Age range: 22-83
Enrolled:102
Evaluation:
male (%): 66
Race: Not stated

Chest pain<12hrs

Trauma or other
explanations for chest pain.
Initial CK level and ECG
results indicative of
myocardial infarction or
unstable angina. Known
tachyarrhythmias.
Uncontrolled hypertension
( 180/110 mmHg).
Significant heart disease.
Known coronary artery
disease.

No definition of AMI.
Follow up for 6 months

AMI:4

Original Articles Summaries

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

New or worsening
wall thickening
abnormalities.

54

Early dobutamine-atropine
stress echocardiography
may safely distinguish
between low-and high risk
subsets for subsequent
cardiac events in patients
with spontaneous chest pain
and non-diagnostic ECG.

90% of patients had

interpretable conditions.

Only one AMI observed for

the study population. Small
numbers prevent clear
conclusions being drawn.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Krasuski, et al,
1999

Exercise testing
using standard and
modified Bruce
protocols

Country: USA
Mean age:56.5
Age range: 25-90
Enrolled: 195
Evaluation: 133
male (%): 55.8
Race: ND

Patients scheduled for

weekend exercise testing

Not stated

30 day follow up: need

for ED visits and
hospitalisation,
6 month follow up:
(MI, cardiac
catheterisation, PTCA,
CABG) and survival.

ND

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Intermediate: ECG
changes 1mm
horizontal or downsloping ST
depression, or chest
pain during exercise
or if patient unable
to achieve adequate
workload
Positive: at least
2mm horizontal or
down-sloping ST
depression or a
1mm ST depression
in conjunction with
typical chest pain,
with a fall in blood
pressure

Unavailable

Unavailable

Weekend and Holiday

exercise testing is a safe and
effective means of risk
stratification prior to
hospital discharge for
patients with chest pain. It
also reduces length of stay
and is cost saving

95.9% (187/195) patients

follow up at 6 months

Insufficient reporting of data

for sensitivity and specificity
calculations

Original Articles Summaries

10

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Lewis, et al,
1999

Immediate exercise
treadmill testing
using modified
Bruce protocol

Country: USA
Mean age:58
Age range:30-85
Enrolled:100
Evaluation:100
male (%):64
Race: Not stated

Patients with confirmed

coronary artery disease.
Low risk patients
determined by ECG and
clinical findings

ECGs diagnostic of AMI or

ACI.
ST-segment and T-wave
changes that would prevent
accurate interpretation of
exercise ECG. Patients with
pulmonary or vascular
disease. Heart failure.
Suspicious aortic dissection.
chest pain believed to be
from non-cardiac origin.

Presence of chest pain

and either abnormal
serial ECG changes of
AMI or cardiac
isoenzymes.

Coronary event: 10

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

1mm horizontal or
down-sloping STsegment depression
or elevation 80 ms
after the J point,
significant
arrhythmia, systolic
blood pressure
decrease of 10mm
Hg or significant
symptoms

Not performed

Not performed

Immediate exercise
treadmill testing of low risk
patients with chest pain and
known CAD is effective in
further stratifying this group
into patients who can be
safely discharged and those
who require hospital
admission.

Poor description of
reference criteria. Poor
description of follow up
procedures

Insufficient reporting to
determine sensitivity and
specificity of the test results

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Sarullo, et al,
2000

Immediate exercise
treadmill testing
using Bruce
protocol

Country: Europe
Mean age: 57
Age range: ND
Enrolled: ND
Evaluation: 190
male (%): 66.8
Race: Not stated

Age >30
Anterior, precordial, or left
later chest pain

Trauma
X-ray abnormalities

Follow up for 12 months.

Chart review, and
telephone interviews.
Endpoints included ED
visits, cardiac events
(AMI, death, PTCA, or
CABG). No definitions
provided.

AMI: 1.1

Original Articles Summaries

11

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

ECG 1mm
horizontal or downsloping ST-segment
depression or
elevation 80ms after
J point. Or
significant
arrhythmia, systolic
BP decrease
>10mmHg, or
significant
symptoms.

84

92.5

Immediate exercise
treadmill testing of low-risk
patients with chest pain who
are at sufficient risk to be
designated for hospital
admission is effective in
further stratifying this group
into those who can be safely
discharged immediately and
those who require
hospitalisation.

No reported loss to
follow up.

Poorly defined criteria for

diagnosis.

Non diagnostic test

classed as peak
heart rate <85% of
predicted

Biomarkers
Single sample measures of CK, CK-MB, cTnI, cTnT, and myoglobin
Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Huggon, et al,
2001

Biochemical
markers measured
at presentation to
ED

Country: UK
Mean age:
Incomplete
Age range: Not
stated
Enrolled:323
Evaluation:227
male (%):
Incomplete
Race: Not stated

Chest Pain or history of MI

Not Stated

Retrospective
examination of medical
records - History of
typical chest pain, serial
ECG showing
evolutionary changes,
serial CK values rising
above twice the upper
limit of normal.

AMI 22.0 (50/227)

Original Articles Summaries

12

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Initial ECG - ST
segment elevation
2mm in two
contiguous chest
leads, or ST
segment elevation
of 1mm in two
contiguous limb
leads.
Troponin T 0.1ng/ml
Myoglobin 90ng/ml
Total CK - 170 U/L
women, 190 U/L
males
CKMB activity
24U/L -CKMB
mass - 5ng/mL then
7.5ng/ml

ECG 70
Troponin T 40
Myoglobin 44
Total CK 40
CKMB activity 33
CKMB mass 52

ECG 99
Troponin T 90
Myoglobin 86
Total CK 86
CKMB activity 90
CKMB mass 88

The ECG is of more

diagnostic use than
biochemical markers in the
first six hours after the onset
of pain, but biochemical
markers give additional
positive diagnostic
information in patients
presenting later than this.
The negative predictive
accuracy of biochemical
markers is too low for a
single sample to be useful
for excluding myocardial
infarction in the first six
hours after onset of
symptoms.

29.7 % (996/3230) of
patients lost to follow up

Questions over the number

and quality of patients
enrolled, as doctors were
requested to take extra blood
samples from possible AMI
patients. Difficult to
determine adherence and the
population from which the
study sample was drawn.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Ooi, et al, 2000

Initial ECG, CKMB, Troponin-T

rapid assay (cTnT)
singly or combined

Country: Asia
Median age: 62
Age range: 27-95
Enrolled:165
Evaluation:152
male (%):67.1
Race:63.8%
Chinese

Age 20
Typical chest pain
Atypical chest pain in whom
AMI cannot be excluded
with out further observation.
No chest pain but with
clinical presentation
suggestive of AMI.

Known MI, or treatment for

MI in last 2 weeks.
Non-cardiac chest pain
Chest trauma.
Cardiac compression or
defibrillation done prior to
blood samples taken.
Recent IM injection
Uraemia clinically or with
raised serum creatine >
150mol/L.

WHO criteria: (1)

clinical history of chest
discomfort >30 min, (2)
appropriate 12-lead ECG
changes; (3) rise and fall
of serum cardiac
enzymes.

AMI: 50

Original Articles Summaries

13

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

cTnT > 0.2ng/mL

CK-MB mass >
6ng/mL
Relative index 5%

ECG: 76.3
cTnT: 31.6
CK-MB mass: 38.2
Combined scores:
89.4

ECG: 79.7
cTnT: 100
CK-MB mass: 97.4
Combined scores: 78.3

The first ECG was the most

sensitive test while the
combination of chest pain,
ECG, cardiac enzymes and
cTnT gave the best results
in the initial diagnosis of
AMI. If the first ECG, CKMB mass and cTnT are all
negative, the probability of
having an AMI is 12%.

4.6% (7/152) patients

lost to follow up.

Very high rate of AMI and

hospital admission (97%).
Serial measurements may
have given more accurate
results.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Porela, et al,
2000

ECG, CK-MB, cTnI

at admission

Country: Finland
Mean age:68
Age range: Not
stated
Enrolled: 311
Evaluation:301
male (%):Not stated
Race: Not stated

Chest pain, dyspnoea, or

arrhythmia suggestive of
acute myocardial ischaemia

Trauma or operation

CK-MB level 11g/L

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

ECG: modified
Aldrich score
greater 10, or
horizontal or down
sloping ST
depression 1mm
CK-MB mass
5g/L
cTnI 0.1g/L

ECG: 66
CK-MB: 60
cTnI: 53

ECG: 65
CK-MB: 92
cTnI: 91

ECG or CK-MB or
cTnI: 90

ECG or CK-MB or cTnI: 54

The commonly available

biochemical and ECG
criteria allow risk
stratification of patients
with a suspected acute
ischemic event. The data
analysis can easily be
automated and is
independent of patient
delay.

Original Articles Summaries

Limitations/comments

14

Serial Measurements of CK, CK-MB, cTnI, cTnT, and myoglobin

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Bock, et al,
1999

Serial total CK-MB,

and isoform
analysis, at 0, 1, &
3 hours post
presentation with
chest pain,

Country: USA
Mean age:58.2
Age range:17-92
Enrolled: 231
Evaluation:231
male (%):65.4
Race: Not stated

Chest pain or other cardiac

symptoms not more than 12
hours prior to presenting to
ED.

Patients immediately
starting thrombolytic
therapy.

WHO criteria, using

ECG and CK-MB mass
at 0, 8 and 16hr after
presentation.

AMI: 16.8

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Total CK-MB :
6IU/L, or CK-MB2
>2.0IU/L or
MB2/MB1 ratio >
1.7, for any sample

68

92

Analysis of CK-MB by
high-voltage electrophoresis
is an effective method for
rapid diagnosis of MI with
the isoform analysis
enhancing early sensitivity.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Bholasingh, et
al, 2001

CK-MB mass
sampling at
admission, 5,7, &
10 hours post onset
of symptoms.

Country: Europe
Mean age:62
Age range: 24-96
Enrolled: 1109
Evaluation: 653
male (%): 56.8
Race: Not stated

Chest pain with typical

cardiac chest pain.

Patients transferred to CCU

(eg evidence of myocardial
damage, unstable angina).
Patients discharge from ED
with atypical or non-cardiac
chest pain.

Follow at 30 days, six

months and 24 months
by telephone interview,
with patient or associated
person. Medical records
examined for cardiac
events using WHO
criteria to define AMI.

ND: Data reported only

patients that were initially
excluded from an AMI
diagnosis.

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

CK-MB mass limit

set at 7g/L for
1996 cohort.

Not Relevant:
clinical impact
study

Not Relevant: clinical

impact study

Using a rule-out AMI

protocol with a rapid and
sensitive CK-MB mass
assay and serial sampling,
the LOS of patients with
chest pain in the cardiac ED
can be reduced without
compromising safety.

CK-MB mass limit

set at 8.0 U/L for
1994 cohort.
Original Articles Summaries

Isoform data not available

until 1 week post admission.
This study raises questions
concerning the feasibility of
using this technique within
ED setting.

Compares two samples over

two different time periods.
Potential for historic events
to bias results.

15

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Fesmire, 2000a

CK-MB and cTnI at

baseline, and 2hr
delta CK-MB and
delta cTnI.

Country: USA
Mean age:54.7
Age range: Not
stated
Enrolled: Not
Relevant
Evaluation: 578
male (%):57.8
Race: 74.2
Caucasian

Chest pain.
Baseline CK-MB 12ng/mL
and cTnI 3ng/mL a second
CK-MB and cTnI at
approximately 2hrs

Recent cocaine use.

Tachyarrhythmia.
Pulmonary edema.

WHO criteria: chest pain

30 minutes, and any of
the following: serial rise
of CK-MB 12ng/mL
and CK-MB index 4%,
a serial rise in cTnI
3ng/mL; new Q wave
formation in two
contiguous leads: patient
death.

AMI: 9.9

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Retrospective
analysis based on
ROC data.
2hr cTnI
0.9ng/mL
2hr Ck-MB
6.4ng/mL
Delta CK-MB:
1.5ng/mL
Delta cTnI
0.2ng/mL

Both 2hr delta CKMB and delta cTnI

had sensitivities of
91.2

Both 2hr delta CK-MB and

delta cTnI had specificities:
91.4

Patients with either a rise in

CK-MB of 1.5ng/mL or a
rise in cTnI 0.2ng/mL in
two hours should receive
consideration for aggressive
anti-ischaemic therapy and
further diagnostic testing
before making an
exclusionary diagnosis of
non-ischaemic chest pain.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Fesmire, 2000b

Continuous 12 lead
ST segment
monitoring with
automated serial
ECG monitoring
(SECG) with
change in CK-MB
from presentation to
ED, and 2hrs later.

Country: USA
Mean age: 54.5
Age range: Not
stated
Enrolled: 706
Evaluation: 706
male (%): 58.6
Race: 77%
Caucasian

Chest pain, with suspected

ischaemic heart disease

Recent cocaine use.

Tachyarrhythmia. Demand
pacemaker. Pulmonary
oedema.

WHO criteria: chest pain

30 minutes, and any of
the following: serial rise
of CK-MB 12ng/mL
and CK-MB index 4%,
a serial rise in cTnI
3ng/mL; new Q wave
formation in two
contiguous leads: patient
death.

AMI: 10.8

Original Articles Summaries

Retrospective analysis..

16

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Base line CK-MB

>6ng/mL
cTnI >1.5ng/mL
Increase of CK-MB
of 1.5 ng/mL from
initial to 2hr test.

Initial ECG: 39.8

Plus baseline CKMB: 55.4
Plus SECG: 74.7
Plus delta 2hr CKMB: 94.0

Initial ECG: 98.7

Plus baseline CK-MB: 98.6
Plus SECG: 97
Plus delta 2hr CK-MB: 93.6

SECG monitoring in
conjunction with the 2 hour
delta CK-MB allows for
early identification and
exclusion of MI, and can
assist the ED physician in
making appropriate
treatment and disposition
decisions.

ECG: if evaluating
physician
interpreted the ECG
as revealing new
injury or new
ischemia.

Potential verification
bias

Limitations/comments
Very promising results

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Herren, 2001

Serial
measurements of
CK-MB mass at 3
and 6 hours post
onset of chest pain,
and continuous ST
segment monitoring
(12 lead) for six
hours

Country: UK
Mean age: Not
stated
Age range: Not
stated
Enrolled: 383
Evaluation: 292
male (%):Not stated
Race: Not stated

Chest pain for less than 12

hours
Age>25

Trauma
Other medical cause of
chest pain
ECG evidence of AMI or
MI
Hypertensive or arrhythmia

48 hours post admission,

either troponin T
concentration (positive if
0.1g/l), or screening for
MI using WHO criteria.

AMI: 12.3 (36/292)

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Change in ST
segment 2mm in
any one lead, or
1mm in any three
leads, Or Absolute
CK-MB mass
concentration >
5g/l or of 3g/l
on consecutive
samples

97.2

93

The six-hour rule-out

protocol for MI is accurate
and efficacious. It can be
used in patients presenting
to ED with chest pain
indicating low to moderate
risk of MI.

23.8% total loss to

follow up, due to refusal
to participate, violation
of protocol or inadequate
follow up.

Important study. Questions

over appropriateness of gold
standard (longer time
frame?) and large proportion
lost to follow up.

Original Articles Summaries

17

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Peacock, et al,
2001

Serial CK and MB
fractions measured
on arrival to ED
then 8 and 16 hrs
post presentation.

Country: USA
Mean age:62.1
Age range: Not
stated
Enrolled: 698
Evaluation: 698
male (%):58.3
Race:57.8
Caucasian

Suspicion of acute coronary

syndrome, based on
emergency physician based
assessment of history, ECG
and physical examination.
Age >18

Unable to obtain informed

consent.
thrombolytic therapy,
coronary revascularization,
or MI within 3 weeks of
presentation,
cardiopulmonary
resuscitation within 7 days
of presentation, major
surgical procedure within 7
days, need for blood
pressure support, transfer
from another hospital,
symptoms present for more
than 12 hours prior to ED
admission, chronic dialysis,
or baseline creatine of >2.0
mg/dL.

WHO criteria; chest pain

consistent with cardiac
origin and new Q-waves
> 0.04 ms.

acute coronary syndrome:

24.9

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

CK < 220U/L or
CK-MB <
8.8ng/mL

Acute 34.5
6 months 32

Acute 88.7
6 months 84

The adverse event rate for

patients with suspected
acute coronary syndromes
and elevated MB is the
same whether or not the
total CK is elevated. These
patients should be
considered as having had an
acute coronary syndrome.

No data on loss to follow

up.

No data on loss to follow up.

Low sensitivity values.

A composite
outcome for the
index visit and at 6
months was
developed. The
composite outcome
was defined as
either death , Qwave MI,or a
revascularisation
procedure (either
PTCA or CABG)

Original Articles Summaries

18

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Johnson, et al,
1999

Two or more
measures of cTnI
during first 24 hrs
of presentation to
ED, as well as CKMB

Country: USA
Mean age: 61
Age range: Not
stated
Enrolled: 1477
Evaluation: 1303
male (%): 52
Race: white 65%,
black 24%

Anterior, precordial or left

lateral chest pain

Trauma.
Abnormalities on chest
radiography.
Less than 2 measurements
of cTnT in first 24 hrs post
presentation to ED

CK-MB >5% or total CK

showing typical rise or
fall.
CK-MB mass >5ng/mL
ECG with pathologic Qwave.
Thrombolytic therapy
over next day required if
new ST elevation
evolved, or occlusion of
infarct related artery

11

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Total CK activity
5%
CK-MB mass
5ng/mL
cTnT > 0.1ng/mL

Total CK activity:
64
CK-MB mass: 98
cTnT: 99

Total CK activity: 98
CK-MB mass: 97
cTnT: 85

In a heterogeneous patient
population seen in the ED
with acute chest pain, cTnT
was similar to CK-MB
(activity and mass assays)
for detection of AMI and
superior to the CK-MB
mass assay as a marker for
major cardiac events early
in the hospital stay among
those who were ruled out
for an AMI.

Original Articles Summaries

Further study is required to

determine how to use this
assay to provide more
appropriate, cost effective
care.

19

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Peacock, et al,
2000

cTnI levels
measured at
baseline 4,8 and 16
hrs post admission.

Country: USA
Mean age: 59.8
Age range: Not
stated
Enrolled: 266
Evaluation: 266
male (%): 47.7
Race: 48.1% white.
47 % black

Age >18 years

Suspected acute coronary
syndrome

Thrombolytic therapy,
coronary revascularization,
or MI within past 3 weeks.
Major surgical procedure in
past 7 days. Vasopressors or
beta-blockers, calcium
channel blockers,
angiotension-converting
enzyme inhibitors, or
nitrates. Symptoms > 12
hrs. Long term dialysis or
CK >176.8g/L.

WHO criteria; chest pain

consistent with cardiac
origin and new Q-waves
> 0.04 ms.

Adverse event: 7.9

AMI: 1

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

cTnT > 0.2g/L

For adverse event at

index visit: 9.5

For adverse event at index

visit: 99.2

For adverse event at

6 months: 0

For adverse event at 6

months: 98.4

Determination of troponin T
levels has a low sensitivity
and high specificity for
predicting outcomes in lowrisk patients evaluated for
suspected acute coronary
syndromes. Study does not
support a strategy of relying
solely on troponin testing
for disposition decisions.

Very low incidence of AMI

in sample.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Van Lente, et al,

1999

Measurement of
troponin T & I and
CK-MB in patients
with suspected
acute coronary
syndrome, at
presentation and at
4, 8,16 hrs
thereafter

Country: USA
Mean age:65
Age range: Not
stated
Enrolled: 255
Evaluation: 255
male (%):58
Race: Not stated

Suspected acute coronary

syndromes, with renal
disease. Two cohorts from a
lager database used to match
Troponin T, and Troponin I.

Cardiopulmonary
resuscitation within 7days,
angiography or thrombolytic
therapy within 3 weeks, or
were given vasodepressors.

WHO criteria, with at

least two of: chest pain
consistent with cardiac
origin, appropriate 12lead ECG changes or
characteristic changes in
CK and CK-MB (total
CK 220 U/L and RI
4%) ascertained
retrospectively.

AMI 18.7

Original Articles Summaries

20

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Not stated. Study

design set to
determine utility of
tests in subpopulation.

Range of values
provided

Range of values provided

The ability of cardiac

troponin T & I to predict
risk for subsequent adverse
outcomes in patients
presenting with acute
coronary syndromes is
reduced in the presence of
renal insufficiency.

Case matched study.

Not a prospective study.

Results should only be
assessed in terms of utility
of troponin T & I and CKMB in patients with or
without renal disease.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Wilcox, et al,
2001

Prediction of
outcome based on
CK and CK-MB,
and troponin I
levels, and clinical
symptoms.

Country: Australia
Mean age:67.1
Age range:16-93
Enrolled:434
Evaluation:424
male (%):55
Race: Not stated

Consecutive patients where

CK-MB was requested.

Not stated

Cardiac and all cause

mortality within 30 days
of presentation

Not relevant

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Not relevant as a
predictive value
study

Not relevant

Not relevant

Compared with CK-MB,

cardiac troponin I more
accurately predicts 30-day
mortality rates in patients
presenting to the ED.
Moreover, troponin I levels
identify additional groups of
patients at increased risk of
death not so identified by
measuring CK-MB values.

10/424 patients lost to

follow up.

Not directly related to

diagnosing AMI

Original Articles Summaries

21

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Green, et al,
2000

Single-sample
myoglobin, CKMB, cTnT, physical
examination

Country: USA
Mean age:60.3
Age range:
Enrolled: 470
Evaluation:396
male (%): 49.5
Race:62% white

Males 30, and Females

40, with any of chest pain,
epigastric pain, unexplained
shortness of breath, new
onset of CHF, acute
pulmonary oedema,
dysrhythmia, ECG change;
Or any age with known
coronary artery disease,
diabetes hyperlipidemia,
cocaine use within 72 hrs.

Non cardiac aetiology

clearly documented by
radiography or other
technical assessment,
symptoms temporally
related to direct trauma,
inability to consent.

Telephone follow up at
14 days following
admission to ED. If not
contactable, a letter was
sent out. Medical records
examined. Death records
searched for those
patients that could not be
followed up.
MI defined by WHO
criteria.

AMI: 9.6

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Myoglobin
110ng/mL

Myoglobin: 28.9
CK-MB: 23.7
cTnT: 23.7

Myoglobin: 91.3
CK-MB: 98.3
cTnT: 94.7

The early prognostic

sensitivity of myoglobin
may allow identification of
some high-risk patients
missed by physician
judgment, CK-MB, and
cTnT. Myoglobin should be
considered for use in the ED
based on both diagnostic
and prognostic abilities.

15.7% of patients
excluded from analysis
(59 patients excluded due
to insufficient sera and
15 due to loss to follow
up).

CK-MB 10IU/L
cTnT .2ng/mL

Original Articles Summaries

22

Other Biomarkers
Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Bayes-Genis, et
al, 2000

D-Dimer levels
measured at
admission to ED.
thrombin-antithrombin
complexes,
prothrombin
fragment 1+2,
activated factor VII,
and fibrinogen were
measured.

Country: Europe
Mean age: 56.9
Age range: Not
stated
Enrolled: 300
Evaluation: 257
male (%): 62
Race: Not stated

Age > 25.

Chief complaint: chest pain
(central or left sided) <6hrs.

Coronary revascularization
within 6 months. Presence
of D-dimer altering
conditions. Conditions
known to alter activity of
coagulation system. Non
consenting.

Prolonged chest pain

occurring at rest
accompanied by STsegment elevation or
depression evolving into
pathological Q-wave or
T-wave inversion
confirmed by CK levels
>180U/L and CK-MB
fraction of more than
twice upper limit.

AMI: 14

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Exploratory
analysis. D-dimer
limit of 500 g/L.
1. Conventional
criteria: Presence
of radiating CP.
2. Positive ECG
(new Q-waves, STsegment elevation
0.1mV; STsegment depression
0.1mV; T-wave
inversion if
deviation 0.1mV
from baseline in 2
or more contiguous
leads. Deviations
80ms after J point
in intervals of
0.5mm. T-wave
deviation
3. Initial
CK>180U/L.

D-dimer: 65

D-dimer: 85

14% (257/300) lost to

follow up

Very promising

D-dimer &
conventional
criteria: 92

D-dimer & conventional

criteria: 97

D-Dimer, an expression of
ongoing thrombus formation
and lysis, is a marker of
substantial incremental
value for the early diagnosis
of acute coronary
syndromes presenting with
chest pain. It adds
independent information to
traditional assessment for
myocardial infarction. DDimer can be incorporated
into clinical decision models
in the ED.

Original Articles Summaries

23

Combination of biomarkers
Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Jurlander, et al,
2000

Serum myoglobin,
CK-MB mass, and
Troponin-T, every 6
hours

Country: Europe
Mean age: 63
Age range:28-88
Enrolled:155
Evaluation:150
male (%):68
Race: Not stated

Suspected AMI.
Chest pain at rest for > 30
min, but < 2 hrs.

No serum samples or 12
lead ECG on admission to
ED

WHO criteria. Typical

history, typical chest
pain, transient elevation
of CK-MB 15IU/L.

AMI: 53.5

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

ECG: ST-segment
elevation 0.1mV
in two limb leads or
0.2mV in two
precordial leads.

86.7

61

Analysis of biochemical
tests in the emergency
department prior to hospital
admission could accurately
identify approximately 20%
additional acute myocardial
infarction patents. The
prognosis of these patients
is poor, and they may be a
target for primary PTCA or
new early initiated
aggressive medical
therapies.

3.2% (5/155) lost to

follow up.

Retrospective analysis.

Propose diagnostic
strategy using cTnT
and CK-MB mass
(no limits defined)

Original Articles Summaries

No clear definition of
biomarker limits on the
diagnostic strategy
proposed

24

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Polanczyk, et al,
1999

CK-MB mass
alone, CK-MB
mass with cTnI if
CK value normal,
CK-MB mass
followed by cTnI if
CK-MB value is
normal and ECG
shows ischemic
changes; both CKMB mass and cTnI;
cTnI alone; all the
above in
combination with
early exercise
testing.

Country: USA
Mean age: Not
stated
Age range: Not
stated
Enrolled:4145
Evaluation: 4145
male (%):
Race: Not stated

acute chest pain

Not Stated

Not stated

11.8

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

CK-MB mass >

5ng/mL,
cTnI >0.4ng/mL

CK-MB mass: 68
cTnI: 55
CK-MB mass and
cTnI: 71

CK-MB mass: 98
cTnI: 93
CK-MB mass and cTnI: 92

Measurement of CK-MB
mass plus early exercise
testing is a cost-effective
initial strategy for younger
patients and those with a
low to moderate probability
of myocardial infarction.
Troponin I measurement
can be a cost effective
second test in higher-risk
subsets of patients if the
CK-MB level is normal and
early exercise testing is not
an option.

No definitions of AMI.

Retrospective analysis.
Important data on costs
associated with various
testing options for detecting
AMI.

Original Articles Summaries

25

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Wu, et al, 1999

Reflex algorithm,
based on
myoglobin, total
CK, CK-MB
isoenzyme, cTnI, at
initial, 4, 8, 12 hrs
post presentation

Country: USA
Mean age: Not
stated
Age range: Not
stated
Enrolled: 101
Evaluation: 101
male (%): Not
stated
Race: Not stated

Chest pain. ECG normal on

non-diagnostic

Not Stated

WHO criteria at
discharge

33.6

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Complicated
diagnostic protocol

100

92.5

Compared to performing
four tests on all patient
samples, the reflex
algorithm would have
reduced the number of
necessary tests from 442 to
130 (71% reduction) for
AMI patients, and 871 to
469 (46% reduction) for
non-MI patients, if
prospectively implemented.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Zimmerman, et
al, 1999

Serial sampling of
CK-MB subforms,
myoglobin, total
CK-MB, cTnT and
cTnI, at 2, 4, 6, 10,
14, 18, 22 hrs after
presentation.

Country: USA
Mean age: 55.3
Age range: Not
stated
Enrolled: 955
Evaluation: 955
male (%): 59.8
Race: Not stated

Age >21 years

Chest pain 15 minutes, of
suspected myocardial
origin, occurring within 24
hrs of presentation.

Not stated

CK-MB mass 7ng/mL

and CK-MB index
2.5%.

AMI: 12.5

Original Articles Summaries

Retrospective evaluation.

26

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

CK-MB mass
7ng/mL
CK-MB >9IU/L.
Myoglobin
>85ng/mL
cTnT >01.ng/mL
cTnI >1.5ng/mL

Based on best result

over 22 hr period.
CK-MB subforms:
96.2, 10 hrs
CK-MB mass: 95.7,
22 hrs
CK-MB: 98.1, 14
hrs
Myoglobin: 86.5,
10 hrs
cTnT: 86.5, 10 hrs
cTnI: 95.7, 18 hrs

Based on specificity at the

time of most sensitive
results.
CK-MB subforms: 90.2, 10
hrs
CK-MB mass: 99.6, 22 hrs
CK-MB: 96.1, 14 hrs
Myoglobin: 90.2, 10 hrs
cTnT: 96.4, 10 hrs
cTnI: 93, 18 hrs

The CK-MB subform assay

alone or in combination
with troponin reliably
triages patients with chest
pain and should lead to
improved therapy and
reduced cost.

Use of related test

criteria for a reference
criterion.

Outlines the changes in

sensitivity and specificity of
biomarkers over time
following admission to ED.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Grzybowski, et
al, 2000

Prediction of death
within 7 days in
patients with
suspected AMI
(retrospective
study). Key
predictors were:
female gender, over
65 years, and high
SBP.

Country: USA
Mean age:66.6
Age range: Not
stated
Enrolled: 291
Evaluation: 244
male (%): 58.6
Race: 58.6% white

Chief complaints: chest pain

or shortness of breath.
Age >18

Only patients with outcome

data were analysed.

Examination of medical
records for death within
7 days of ED arrival, MI,
unstable angina, and
cardiac arrest.

AMI - 96%

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Not stated

86

53

A triage rule based on

multivariate model can
identify the group at high
risk of early cardiac death.
This decision rule needs to
be prospectively validated.

Original Articles Summaries

Retrospective study.
Predictor variables need to
be accessed prospectively.

27

Technetium-99m Sestamibi Myocardial Perfusion Imaging

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Kontos, et al,
1999

CT Sestamibi
imaging and serial
myocardial marker
measurement of
CK-MB, total CK
activity, cTnI over 8
hrs.

Country: USA
Mean age:56
Age range: Not
stated
Enrolled: 721
Evaluation: 620
male (%):47
Race: Not stated

Suggestive symptoms of
myocardial ischemia.

Patients at high risk of

coronary disease.
Patients who had had a
PCTA within previous 6
months.

MI within one week of

admission, performance
of revascularization, or
significant coronary
disease on angiography
within 6 weeks.

MI: 9

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Discrete perfusion
defect with
associated
abnormalities in
wall motion and/or
thickening, or Serial
cTnI 2.0ng/mL, or
Serial cTnI
1ng/mL

Perfusion imaging:
92

Perfusion imaging: 67

Early perfusion imaging and

serial cTnI have comparable
sensitivities for identifying
myocardial infarction.
Perfusion imaging identified
more patients who
underwent revascularization
or who had significant
coronary disease, but had
lower specificity. The two
tests can provide
complementary information
for identifying patients at
risk for ACS.

3 patients had
uninterpretable images.
No description of follow
up techniques used to
confirm diagnosis.

Unclear follow up
procedures.

Original Articles Summaries

Serial cTnI
2.0ng/mL: 90
Serial
cTnI1ng/mL: 97

Serial cTnI 2.0ng/mL:96

Serial cTnI1ng/mL: 94

28

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Kosnik, et al,
2001

Technetium-99m
sestamibi imaging
within 2hrs of onset
of pain, in
conjunction with
ECGs, cardiac
enzymes, standard
cardiac rhythm
monitoring.

Country: USA
Mean age:56
Age range: Not
stated
Enrolled: 69
Evaluation: 69
male (%): 43
Race: 26% white

Age18 years
Non diagnostic ECG

Trauma; Arrhythmia
Heart failure; Shock
Pain resolved under 2hrs
Pregnancy; Previous MI;
Current cardiac
complications;
Diagnostic ECG ( 1mm ST
segment elevation or
depression or Q waves).

Not stated

Not stated

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Interpretation by
expert physician
(EP) as positive,
negative, or
equivocal.

Not relevant as
clinical impact
study

Not relevant as clinical

impact study

Sestamibi scanning results

appropriately affected the
EPs estimates of the
probability of AMI and UA
and improved disposition
decisions. Scanning all
low-risk patients would be
likely to have increased
costs.

No reporting on how
follow up was achieved
to determine if a cardiac
event occurred.

Low sample size.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Mariani, et al,
1999

99mTc-GLA
injection, and
imaging 3 hrs latter,
serial CPK levels
on at least 5
occasions.

Country: Europe
Mean age:62
Age range: ND
Enrolled: 28
Evaluation: 28
male (%): 79
Race: Not stated

Highly suggestive
symptoms of AMI.
(Prolonged chest pain
unresponsive to nitrates,
associated with persistent
ST segment changes on
ECG.

Not stated

Review of cardiologist,
based on serial ECGs,
presence of wall motion
abnormalities on
Echocardiogram and on
serial CPK values.

AMI 82

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Independent
evaluation by 3
physicians, who
were unaware of
other diagnostic
information.

Not performed

Not performed

99mTc-GLA localises in
zones of acute myocardial
necrosis when injected
within 9 hrs of onset of
infarction.

Original Articles Summaries

Small number of subjects.

Diagnostic test
characteristics need to be
determined.

29

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Paventi, et al,
2001

Echocardiography,
myocardial marker
analysis, and
myocardial
perfusion imaging
with 99mT
sestamibi at 4 hrs
post presentation.

Country: Italy
Mean age: Not
stated
Age range: Not
stated
Enrolled: 555
Evaluation: 555
male (%): Not
stated
Race: Not stated

Patients with moderate risk

of myocardial ischemia
(typical symptoms lasting
>30 minutes).
Patients with low risk of
Myocardial ischemia
(typical symptoms lasting
<30 minutes, or longer
atypical symptoms).

Inadequate
echocardiographic views.

WHO criteria: at least

two of: symptoms
consistent with
myocardial ischemia for
>30 minutes; evolving
diagnostic ECG changes;
CK-MB by mass assay
>8.0ng/dL with RI >4.
Long term follow up via
telephone interviews, and
death certificate records.

Not stated

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Echocardiography:
hypokinesis or a
kinesis of 2 or more
contiguous
segments, abnormal
wall motion of one
segment visible in 2
different views, or
global hypokinesis
with moderate to
severe systolic
dysfunction
(ejection fraction
>40%) without
segmental wall
motion
abnormalities.

Echocardiography:
100

Echocardiography: 84

Myocardial perfusion
imaging and twodimensional
echocardiography are highly
concordant when performed
on patients initially
considered at low or
moderate risk for MI or
ischemia. When combined
with the clinical
characteristics, both
imaging techniques could
help accurately identify the
patient at high risk who
requires admission and the
patient at low risk who
could undergo early stress
testing to be safely
discharged from the ED.

No clear reporting of
follow up data on
coronary events.

Perfusion imaging:
Discrete perfusion
defect with an
associated
abnormality in wall
motion or
thickening.

Original Articles Summaries

Perfusion imaging: 86
Perfusion imaging:
100

30

Electron beam computed tomography

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Laudon, et al,
1999

Electron-beam
computed
tomography
(EBCT) of coronary
arteries plus other
cardiac tests

Country: USA
Mean age: 47.9
Age range:30-65
Enrolled: 105
Evaluation: 100
male (%):54
Race:93% white

Men <55, Female < 65

Normal initial cardiac
enzyme values, and normal
or indeterminate initial
ECG.

Known coronary artery

disease, known AMI,
hemodynamic instability,
pregnancy

Treadmill exercise test,

radionuclide stress test,
exercise echocardiogram
interpreted as positive by
cardiologist

AMI:14

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Review by
cardiovascular
radiologist with the
aid of automated
software to identify
coronary artery
calcification.

100

63

EBCT is a rapid and

efficient screening tool for
patients admitted to the ED
with angina-like chest pain,
normal cardiac enzyme
concentrations,
indeterminate ECG findings
and no history of coronary
artery disease. Such patients
with negative EBCT results
may be safely discharged
from the ED without further
testing or observation.

Small numbers tested, but

very promising results

Goldman Chest Pain Protocol

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Durairaj, et al,
2001

Goldman prediction
rule

Country: USA
Mean age:57.5
Age range:17-98
Enrolled:1061
Evaluation: 1033
male (%):48.7
Race:27% African
American

Chest pain, and non chest

pain with indicators of
cardiac origin of symptoms.

Patients admitted from

outpatients, transfers from
inpatient services or other
hospitals were excluded.

Cited previous work.

2.9%

Original Articles Summaries

31

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Not stated, cited

previous work

Based on ruling out

very low risk
patients only: 76.7.

Based on ruling out very

low risk patients only: 52.3.

The prediction rule

accurately identified
patients with or without
chest pain who were at very
low risk of major
complications, identifying a
subset from whom cardiac
monitoring could be
withheld safely.

2.6% (28/1061) Patients

lost to follow up.

Follow up only over 72 hour

period.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Reilly, et al,
1999

Goldman prediction
rule

Country: USA
Mean age: Not
stated
Age range: Not
stated
Enrolled: 215
Evaluation: 207
male (%): 45
Race: not stated

Suspected acute ischaemic

heart disease.
Age > 18 years.

Not stated

Not stated

Major cardiac
complications: 4.3

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

ECG suspicious for

AMI, ECG
suspicious for acute
ischemic, systolic
blood pressure <
100 mm Hg, rales
bilaterally above the
lung bases, known
history of unstable
ischemic heart
disease.

Not performed.

Not performed.

This independent validation

of the prediction rule
suggests that it can improve
triage decisions for patients
admitted with suspected
acute ischemic heart
disease. Additional studies
are needed to test
prospectively the
performance of the
prediction rule in actual
decision making, its
acceptance by clinicians and
its cost effectiveness.

Original Articles Summaries

Unclear how independent

diagnosis was achieved, or
the method of follow up
undertaken. As such, results
are very difficult to interpret.

32

Other Computer-Based Decision Aids

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Aase, 1999

Decision support
computer program
(DSP) in addition to
clinical judgements,
and ECG findings

Country: Norway
Mean age: Not
stated
Age range: Not
stated
Enrolled: 493
Evaluation: 493
male (%):64.3
Race: Not stated

Chest pain

Not stated

Not stated

AMI: 36.1

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Decision made by a
computer program
using Bayes
theorem

84.3

75.2

Use of DSP in the

emergency room on easily
available anamnestic and
clinical variables may
improve referrals to the
CCU, optimise therapy and
resource use.

No information on how
follow up and final
diagnosis was
determined.

Left bundle-branch blocks sub-population specific studies

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Kontos, et al,
2001a

Clinical histories,
ECG and serial CK
measurement in ED
of patients with left
bundle-branch
block.

Country: USA
Mean age:66
Age range: Not
stated
Enrolled: 7725
Evaluation:182
male (%):30.2
Race: Not stated

Left bundle-branch block

(LBBB)

Not stated

CK-MB 8.0 ng/mL and

Relative index = 4.0 and
characteristic increase
and decrease in markers.

AMI: 13

Original Articles Summaries

33

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Retrospective
analysis of records

New LBBB: 42
Combined STsegment and CKMB: 63

New LBBB: 42
Combined ST-segment and
CK-MB: 99

ECG criteria for identifying

patients with AMI and
LBBB identify only a small
minority of patients with
AMI. Treating all patients
with LBBB and chest pain
with fibrinolytics would
result in treatment of a
significant number of
patients without AMI.

ND

Unclear how follow up was

achieved. Important work on
evaluating AMI in a subpopulation.

Study, year,

Diagnostic
protocol

Location,
population

Inclusion criteria2

Exclusion criteria

Reference criteria

Prevalence
of disease (%)

Shlipak et al.,
1999

Retrospective
analysis of initial
ECG data to predict
AMI in patients
with LBBB.

Country: USA
Mean age: ND
Age range: ND
Enrolled: 83
Evaluation: 103
admissions
male (%): ND
Race: ND

Age > 18 years.

Acute cardiopulmonary
symptoms and complete
LBBB on initial 12-lead
ECG.

No measure of CK or
troponin I within 12 hrs of
initial ECG.

MI: characteristic
clinical presentation and
troponin I 1.5 mg/L or
CK-MB fraction 7U/L

AMI: 30

Test criteria for

AMI

Sensitivity (%)

Specificity (%)

Study conclusions

Potential verification
bias

Limitations/comments

Any 3 ST-segment
abnormalities

23

82

The ECG is a poor predictor

of MI in a community-based
cohort of patients with
LBBB and acute
cardiopulmonary symptoms.
Acute thrombolytic therapy
should be considered for all
patients with LBBB who
have symptoms consistent
with MI.

Multiple admissions into

the study were allowed.
The authors analysed the
possible effects of this,
and found that the results
did not change when the
subsequent admissions
were omitted.

Questionable use of
readmitting the same patient
on multiple ED admissions.

Original Articles Summaries

34

Appendix 1: Summary of guidelines concerning AMI

Summary 1
American College of Emergency Physicians. Clinical Policy: Critical Issues in
the Evaluation and Management of Adult Patients Presenting with Suspected
Acute Myocardial Infarction or Unstable Angina
Ann Emerg Med. May 2000;35:521-544.
This clinical policy focuses on critical issues in the evaluation and management of patients
with acute myocardial infarction or unstable angina. A MEDLINE search for articles
published between January 1993 and December 1998 was performed using combinations of
the key words chest pain, acute myocardial infarction, unstable angina, thrombolytics,
primary angioplasty, 12-lead ECG, ST-segment monitoring, cardiac serum markers, and chest
pain centers. Subcommittee members and expert peer reviewers also supplied articles with
direct bearing on the policy.
This policy focuses on 5 areas of current interest and/or controversy: (1) ECG eligibility
criteria for fibrinolytic therapy, (2) role of primary angioplasty in patients with acute
myocardial infarction, (3) use of serum markers to diagnose acute myocardial infarction, (4)
serial 12-lead ECGs during the initial evaluation, and (5) chest pain evaluation units.
Recommendations for patient management are provided for each of these 5 topics based on
strength of evidence (Standards, Guidelines, Options). Standards represent patient
management principles that reflect a high degree of clinical certainty; Guidelines represent
patient management principles that reflect moderate clinical certainty; and Options represent
other patient management strategies based on preliminary, inconclusive, or conflicting
evidence, or based on panel consensus. This guideline is intended for physicians working in
hospital-based emergency departments or chest pain evaluation units.
Methodology
This clinical policy was created after careful review and critical analysis of the peer-reviewed
literature. A MED-LINE search for articles published between January 1993 and December
1998 was performed using combinations of the key words chest pain, AMI, unstable angina,
and thrombolytics. Abstracts were reviewed by subcommittee members, who then selected
the following topics on which to focus this policy: (1) ECG eligibility criteria for fibrinolytic
therapy, (2) role of primary angioplasty in patients with AMI, (3) use of serum markers to
diagnose AMI, (4) serial 12-lead ECGs during the initial evaluation, and (5) chest pain
evaluation units. Additional MEDLINE searches were performed using the key words 12lead ECG, ST-segment monitoring, cardiac serum markers, and chest pain centers. Pertinent
articles were selected from the reviewed abstracts and from bibliographies of initially
selected papers. Committee members and expert reviewers also supplied papers from their
own knowledge base. All publications were stratified by at least 2 of the subcommittee
members into 1 of 3 categories of strength of evidence, and some were down-graded 1 or
more levels as necessary based on a standardized formula that graded papers on size,
methodology, validity of conclusions, and potential sources of bias.
This policy is not intended to be a complete manual on the initial evaluation and management
of patients with AMI and unstable angina. Some areas suggested by expert peer reviewers for

Appendix 1: Guidelines Summaries

35

addition of further discussion included utilization of serum markers to risk stratify unstable
angina patients, use of combinations of serum markers to exclude AMI,risk stratification
tools such as the acute time-insensitive predictive instrument (ACI-TIPI) and Goldman
criteria for predicting need of intensive care admission, and discussion of multiple
technologies for identifying acute coronary syndromes (ACS). These areas have been
discussed to some degree in other clinical policies and represent areas that ACEP may
address in future updates of this current policy. The reasons for developing clinical policies in
emergency medicine and the approaches used in their development have been enumerated.
This policy is a product of the ACEP clinical policy development process, including expert
review, and is based on the existing literature; where literature was not available, consensus
of emergency physicians was used. Expert review comments were received from emergency
physicians, physicians from other specialties, such as cardiologists, and specialty societies
including members of the American Academy of Family Physicians, American Association
for Clinical Chemistry, and the American Society of Nuclear Cardiologists. Their responses
were used to further refine and enhance this policy. Clinical policies are scheduled for
revision every 3 years; however, interim reviews are conducted when technology or the
practice environment changes significantly. During the review process, all papers were given
a baseline strength of evidence by the subcommittee members, according to the following
criteria:
Strength of evidence AInterventional

studies including clinical trials, observational studies

including prospective cohort studies, aggregate studies including meta-analyses of
randomized clinical trials only.
Strength of evidence BObservational

studies including retrospective cohort studies, casecontrolled studies, aggregate studies including other meta-analyses.
Strength of evidence CDescriptive

cross-sectional studies, observational reports including

case series, case reports; consensual studies including published panel consensus by
acknowledged groups of experts. Strength of Evidence A and B papers were then rated on
elements the committee believed were most important in creating a quality work. A and B
papers with significant flaws or design bias were downgraded from 1 to 3 levels based on a
set formula. Strength of Evidence C articles were downgraded 1 level if they demonstrated
significant flaws or bias. Articles downgraded below a C strength of evidence were given
an X rating and were not used in formulating this policy. Clinical findings and strength of
recommendations regarding patient management were then made according to the following
criteria:
Evidence-based standards. Generally accepted

principles for patient management that reflect a

high degree of clinical certainty (ie, based on strength of evidence A or overwhelming
evidence from strength of evidence B studies that directly address all the issues).
Guidelines. Recommendations

for patient management that may identify a particular strategy

or range of management strategies that reflect moderate clinical certainty (ie, based on
strength of evidence B that directly addresses the issue, decision analysis that directly
addresses the issue, or strong consensus of strength of evidence C).
Options. Other

strategies for patient management based on preliminary, inconclusive, or

conflicting evidence, or, in the absence of any published literature, based on panel consensus.

Appendix 1: Guidelines Summaries

36

Scope of Application
This guideline is intended for physicians working in hospital-based emergency departments
or chest pain center evaluation units.

Patient management recommendations

1.

ECG eligibility criteria for emergent reperfusion therapy

Evidence-based standards. Assess

for fibrinolytic therapy in patients presenting within 12

hours of symptom onset if ECG reveals:
1. ST-segment elevations greater than 0.1 mV in 2 or more contiguous leads that are
not characteristic of early repolarization or pericarditis, nor of a repolarization
abnormality from LVH or BBB in patients with clinical presentation suggestive
of AMI.
2. Any type of BBB (right, left, paced, and atypical new or old) in patients with
clinical presentation suggestive of AMI.
Guidelines. Assess

for fibrinolytic therapy if ECG reveals LBBB and ST-segment deviations

of 1 mm or more toward the major QRS deflection or 5 mm or more away from the major
QRS deflection in 2 or more contiguous leads in patients with atypical presentation of AMI.
Options. Assess

1.
2.
3.

2.

for fibrinolytic therapy if ECG reveals:

ST-segment depressions of 1 mm or more with upright T waves in 2 or more
contiguous anterior precordial leads in patients with clinical presentation
suggestive of pos-erior AMI.
ST elevations of 1 mm or more in 2 or more contiguous nonstandard leads (V 4 R
through V 6 R, V 7 through V 9 ) in patients with clinical presentation suggestive
of isolated right ventricular or posterior AMI.
RBBB, atypical BBB, or paced BBB and ST-segment deviations of 1 mm or
more toward the major QRS deflection or of 5 mm or more away from the major
QRS deflection in 2 or more contiguous leads in patients with atypical
presentation of AMI.

Role of primary angioplasty in patients with AMI

Evidence-based standards. Primary coronary angioplasty when

performed by experienced
personnel within 90 minutes of diagnosis of AMI is as effective as fibrinolytic therapy in
AMI patients meeting standard criteria for emergency reperfusion therapy.
Guidelines. If

resources are available, consider primary coronary angioplasty as an alternative

to fibrinolytic therapy in AMI patients meeting standard criteria for emergent reperfusion
therapy providing it can be performed within 90 minutes of diagnosis of AMI.
Options. None

3.

specified.

Serum marker analysis in AMI

Evidence-based standards. No

single determination of one serum biochemical marker of

myocardial necrosis reliably identifies* or reliably excludes** AMI less than 6 hours of
symptom onset. No serum biochemical marker identifies or excludes unstable angina at any
time after symptom onset.
Guidelines. In patients presenting with acute chest pain and a negative baseline serum marker

Appendix 1: Guidelines Summaries

37

level, consider repeat serum marker testing at the following time intervals from symptom
onset before making an exclusionary diagnosis of non-AMI chest pain:

CK-MB activity 812 hours

CK-MB mass 610 hours

CK-MB subforms 610 hours

cTnT 812 hours

cTnI 812 hours

The exact timing of the repeat determination of the serum marker value should take into
account the sensitivity, precision, and institutional norms of the assay being used, as well as
the release kinetics of the marker being measured. CK-MB activity, CK-MB mass, cTnT, and
cTnI all reliably identify and exclude AMI 12 to 24 hours after symptom onset. Because of its
rapid release kinetics, myoglobin alone does not reliably identify or exclude AMI at any time
interval after symptom onset and is best used in conjunction with the other common serum
markers. cTnT and cTnI are the preferred serum markers in patients presenting greater than
24 hours after symptom onset.
Options. Consider

repeat determination of CK-MB mass 2 to 3 hours after baseline or repeat

myoglobin at 1 to 2 hours after baseline for utilization of the .CK-MB or .myoglobin when
the repeat serum marker level is drawn at a time interval before the time intervals discussed
in the Guidelines recommendation above.
*Reliably identifies = sensitivity >=90% with positive likelihood ratio >=10
**Reliably excludes = specificity >= 90% with negative likelihood ratio <= 0.1

4.

Serial 12-lead ECGs in the ED

Evidence-based standards. Performing

SECG or repeat ECGs at select time intervals after

presentation results in an incremental increase in identification of injury or ischemia in
patients with AMI and unstable angina compared with the baseline ECG. Its greatest value
appears to be when it is used in patients with intermediate or high clinical likelihood of AMI
or unstable angina who are spending at least 1 hour in the ED or in identification of
successful reperfusion from fibrinolytic treatment.
Guidelines. Perform

repeat ECG at a set time interval after presentation or automated SECG

monitoring during the ED evaluation of patients in whom the initial ECG is non-diagnostic
for injury and who have symptoms consistent with ongoing or recurrent ischemic chest pain.
Options. Perform

repeat ECG at a set time interval after presentation or automated SECG

monitoring during the ED evaluation of patients with a low suspicion of AMI or unstable
angina.
5.

Chest pain evaluation units

Evidence-based standards. Chest

pain evaluation units are a safe and effective alternative to

routine admission for evaluation of low- to intermediate-risk chest pain patients. Further
investigation needs to be performed to determine the most cost-effective and efficient
utilization of available diagnostic modalities.
Guidelines. As an alternative to admission, consider use of a CPEU protocol consisting of
serial serum marker determinations, serial ECGs, and selective stress testing for evaluation
and risk stratification of patients at low- to intermediate-risk for AMI and ACS.
Options. None stated.
Appendix 1: Guidelines Summaries

38

Summary 2
CHEST PAIN: CLINICAL DECISION-MAKING
American College of Emergency Physicians,
1998 Scientific Assembly
This document provides a guide to all causes of chest pain and to the diagnostic tools which
are appropriate for use with each of the causes identified.
A

Diseases You Can't Afford to Miss

1. Myocardial infarction
WHO Criteria: At least 2 of the following:
(1) Ischemic chest discomfort > 30 min
(2) ECG evolution: ST waves, Q waves
(3) Serum cardiac marker changes
2. Unstable Angina
Presentation
(1) New onset of chest pain > 20 mins
(2) Chest pain at rest > 20 mins
(3) Chest pain at previously tolerated levels of activity
(4) Chest pain not relieved by previously effective dose of nitroglycerin
3. Aortic Dissection
Presentation
(1) Sudden, excruciating chest pain
(2) Often "tearing" and radiating to back
(3) May have neurologic abnormalities, CHF, syncope
(4) May have BP differential between arms
(5) May have murmur of AI
4. Pericarditis
Presentation
(1) Sharp, central chest pain
(2) Pain worse recumbent; better leaning forward
(3) Pericardial friction rub is pathognomonic (LLSB)
(4) ECG findings
(a) Diffuse concave-upward ST segment elevation
(b) Sometimes PR-segment depression
5. Cardiac Tamponade
Presentation
(1) Chest pressure/discomfort with SOB
(2) May see elevated JVP, hypotension, tachypnea, narrow pulse pressure, pulsus
parodoxus
(3) May see electrical alternans on ECG
(4) Echocardiography is diagnostic (ED or formal US)
6. Pulmonary Embolus
1. Presentation
(1) 96% will have either dyspnea, pleuritic pain or tachycardia
Appendix 1: Guidelines Summaries

39

(2) Sx: Chest pain (88%), dyspnea (84%), apprehension/anxiety (59%), syncope
(13%)
(3) Findings: RR > 16 (92%) [but who counts, anyway], rales (58%), HR > 100
(44%), Temp > 37.8 (43%)
2. Diagnostic Modalities:
(1) CXR - Usually non-contributory
(2) ECG - May see sinus tachycardia or rarely S1-Q3-T3
(3) V/Q Scan - 85% Sensitive. Must interpret in light of your pre-test probability.
(4) Pulmonary angiogram - Still the gold standard: 96-98% sensitive and specific
(5) Helical CT - Becoming more popular. Sensitivity close to V/Q. Will it replace it?
(6) MRI - Under evaluation. May prove quite useful.
(7) Lower extremity doppler ultrasound - Helpful if positive. Not helpful if negative.
(8) D-Dimer - Degradation product of fibrin. Disagreement about utility at this time.
(9) ABG - 10-15% have normal A-a gradient. Little utility unless pre-test probability
is very low.
7. Pneumothorax
8. Acute Chest Syndrome of Sickle Cell Disease
Adult presentation
(1) Chest pain: 84% (Severe, 44%)
(2) Fever: 64%
(3) Cough: 63%
(4) Shortness of breath: 47%
(5) Chills: 39%
(6) Diagnosis made by new infiltrate on CXR
9. Esophageal Rupture
Presentation
(1) Severe chest pain, usually pleuritic
(2) Dyspnea, SOB, cyanosis
(3) Odonophagia
(4) Pain on neck flexion
(5) Pale, diaphoretic, severely ill appearing
(6) Subcutaneous emphysema
(7) Hamman's crunch
(8) CXR: May see pneumo-mediastinum, pleural effusion, pneumothorax, widened
mediastinum
If undiagnosed, is uniformly fatal
B. Diseases You Would Like Not to Miss
1. Pneumonia
2. Esophageal Disease (Spasm, Reflux, Inflammation)
3. Peptic or Gastric Ulcer (with or without perforation)
4. Cholecystitis
5. Early Disseminated Lyme Disease
May present with chest pain, syncope, dizziness, SOB, A-V block

Appendix 1: Guidelines Summaries

40

C. Benign Diagnoses (Relatively)

1. Panic Disorder
(1) Intense fear accompanied by chest pain, dyspnea, nausea
(2) Seen in 30-50% of patients with chest pain and normal coronary arteries
2. Depression
3. Herpes Zoster
4. Hyperventilation
5. Chest Wall Pain / Costochondritis
D. Making the Diagnosis of Myocardial Infarction
1. History
(1) Worry more: males, older, h/o MI, pain > 1 hour, diaphoresis
(2) Worry less: "sharp" pain, age < 40
2. Physical Examination
(1) Worry more: rales
3. ECG
(1) (1) 50-60% of AMI patients will have diagnostic changes
(2) Normal or non-diagnostic ECG -> low risk patients
(3) Comparison with old ECG may not change admission decision, but may change level
of acuity
(4) Abnormalities:
ST-T elevation > 1mm in standard pattern
Q waves (.04 sec, 1/3 height of R wave) (except in III)
T wave inversion (unless only in III or V1)
Hyper-acute T waves in 2 or more leads
LBBB (even if old), in light of ischemic pain should be treated aggressively
Suspect RV infarction with distended neck veins & clear lungs:
ST-T elevation of >1mm in V4R is diagnostic
(5) Increased sensitivity through automated serial 12-lead ECG
4. Serum Markers
(1) CK - not used by itself
(2) CK-MB total activity (IU/ml)
(3) CK-MB mass (ng/ml) - The current gold standard CK-MB Index (CK-MB mass /
CK total activity)
(4) CK-MB Subforms - In development
(5) Myoglobin - Sensitive but not specific; narrow window
(6) Troponin-T - Very sensitive; may help risk-stratify unstable angina
a. Troponin-I - Most specific for cardiac injury
b. Combining markers, eg myoglobin/CK-MB mass
c. Delta measurements: Ongoing work to evaluate utility of rise in marker value
while still in normal range
2. Problems with comparing serum marker studies
(1) Assay under evaluation often used as dx criterion of AMI
(2) Inconsistencies in cut-off values used for marker
(3) Wide variation in populations studied
(4) Few studies are outcome-based
5. Neural Networks
1. Network is trained

Appendix 1: Guidelines Summaries

41

2. Network establishes its own decision basis

3. Feedback over time can improve decision process
4. Significant potential
5. Shortcomings:
(1) Individual diagnosis based
(2) Each system a different interface
(3) Feedback of local data probably required
6. Ancillary Studies
1. Exercise Tolerance Test (ETT)
(1) Least expensive
(2) Most available
(3) Can be utilized as part of a "rapid r/o" protocol
(4) Only 65-70% sensitivity for CAD
2. Stress Echocardiography
(1) Intermediate expense
(2) Operator dependent
(3) About 80% sensitive for CAD
(4) Availability?
3. Thallium Perfusion Scan
(1) Short window due to rapid wash-out
4. Sestamibi
(1) Better image, 1-2 hour window
(2) May be as sensitive as stress echo but needs more study
7. Special Cases
1. The Elderly
2. Cocaine-Chest Pain
E. Management Plan for Cardiac Chest Pain
A. Risk Stratification
1. High Risk
(1) Ischemic or new ECG findings
(2) Ongoing chest pain
(3) Abnormal vital signs
(4) Rales
2. Moderate Risk
(1) Nondiagnostic ECG
(2) Symptoms > 20-30 minutes, now resolved
3. Low Risk
(1) Normal ECG
(2) Symptoms < 20-30 minutes, now resolved
B. Evaluation, including EC G must be prompt
C. Infarcting
1. Have a standard procedure
2. Lytics: Fair evidence: up to 6 hrs, delay increases infarct size
3. Angioplasty
D. Who to admit where?
1. High risk: CCU
2. Medium/Low risk:
Telemetry/Chest Pain Observation

Appendix 1: Guidelines Summaries

42

(1) Immediate Exercise Testing in Low Risk Patients

(2) Immediate Myocardial Perfusion Imaging in Low Risk Patients
E. ED Chest Pain Evaluation Units
1. Pros
(1) More rapid throughput of low risk chest pain patients
(2) More cost-effective evaluation of same
(3) An additional alternative to admission or discharge
(4) Additional service line for ED
(5) Potential for improved patient satisfaction
2. Cons
(1) Additional responsibility and effort for emergency physicians
(2) Potential for lost revenue for hospital
(3) Potential for higher charges for patients admitted to hospital
(4) After evaluation unit stay
(5) Potential requirement for capital improvement and increased
(6) Staffing
(7) Potential for inappropriate use as disposition
3. Requirements
(1)Interdisciplinary planning
(2) Location, monitoring & resuscitation equipment
(3) Appropriate staffing
(4) Protocol driven
F. Documentation of the Chest Pain Visit
A. The Minimum (ACEP, 1995)
1. Character of pain
2. Age
3. Associated symptoms
4. Past history
5. Vital signs
6. Cardiovascular exam
7. Pulmonary exam
B. When MI is in the Differential
1. Pain: Quality, onset/duration, location, radiation, made worse by, made better by
2. Associated symptoms: Diaphoresis, nausea/vomiting, SOB/DOE, fatigue, syncope
3. Past cardiac history, risk factors
4. Vital signs
5. Cardiovascular exam: JVD, heart (rhythm, murmurs, gallops)
6. Pulmonary exam (rales, wheezes)
7. Extremities: peripheral edema
8. ECG, CXR
C. Other additional documentation based on diagnostic suspicion
D. The Discharged Chest Pain Patient
1. Are your discharge instructions adequate?
2. Is your record completed at the time of discharge?
3. Re-read your record in your head from the witness stand
(1) Is your record consistent with the chief complaint, the triage
(2) note and the nurses notes?
(3) Does your record support your diagnosis and disposition?

Appendix 1: Guidelines Summaries

43

(4) "Not documented, not done"

a. Does your record address those things we can't afford to miss?
b. If you don't think it's cardiac (or other problems you can't afford to miss),
say so
4. How does your ED deal with patient phone calls?
5. We lose the most $ on the youngest MI's
E. Discharge against medical advice (AMA) patients require special care and documentation
1. Don't get into an ego-duel
2. Provide the best treatment and follow-up you can
3. Explain the life threat
4. Carefully document the refusal of care
5. If you have an AMA form, fill it out completely
6. Use a second person witness; family member is best
7. Always offer future care

Appendix 1: Guidelines Summaries

44

Summary 3
Clinical policy for the initial approach to adults presenting with a
chief complaint of chest pain, with no history of trauma
American College of Emergency Physicians, 1995
Chest pain is one of the most common and complex symptoms for which patients seek
emergency medical care. For these reasons the chief complaint of chest pain was selected as
the focus for the first clinical policy by the American College of Emergency Physicians. The
original policy, published in 1990, was primarily a consensus document based on standard
textbooks, review articles, and clinical practice experience. There was at that time little
scientific evidence in the medical literature to support specific historical or clinical findings
or therapeutic modalities. The original policy was therefore intentionally nondirective on
many issues of specific diagnostic methods and therapy.
Since 1990 there has been an explosion of research and publication activity relating to the
diagnosis and treatment of disorders causing chest pain. Among these are multinational,
multicenter and independent investigator studies on the use of specific therapeutic agents and
diagnostic modalities in coronary artery disease, myocardial infarction, and thromboembolic
disease.
This clinical policy was developed by the ACEP Clinical Policies Committee and went
through ACEPs clinical policy development process, including expert review and field
testing. Expert reviewers included emergency physicians, physicians from other specialties,
and various specialty societies. Comments were received from members of the American
Academy of Family Physicians, the American College of Cardiology, the American Heart
Association, the Internal Medicine Center to Advance Research and Education (IMCARE)
Practice Guidelines Network, and the Emergency Nurses Association. Field test volunteers
represented varied geographic areas, practice settings, and patient populations. All of the
expert review and field test comments and recommendations were given careful
consideration by the Clinical Policies Committee as they made decisions about the policy.
The revised clinical policy on chest pain, like the original in format and tone, concentrates on
identification of situations associated with high morbidity and mortality. There has been an
ongoing effort to balance guidelines between the needs of the emergency physician practicing
in a small hospital and those of the attending emergency physician at a large academic center.
This revision places a greater emphasis on identification of risk factors for specific diseases
and is more directive regarding specific diagnostics and therapies that are no longer
controversial and have been accepted as efficacious. ACEP, however, clearly recognizes the
ultimate importance of the individual clinicians judgment, and that there are variations in
practice.
There remain many evolving issues in diagnosis and therapy awaiting further research before
firm recommendations can be made.
From this policy document the following form has been taken as a good example of all the
factors needing consideration when a patient presents at ED with chest pain.

Appendix 1: Guidelines Summaries

45

Quick Reference Form: Chest Pain Revision

For adults with chest pain and no history of trauma. This policy does not include history of
proximate trauma; or visible lesions; or isolated breast disease; or pain of very brief duration
(lasting for seconds only).
Circle line number if yes. Bolded actions are rules. Actions not bolded are guidelines.
Chief Complaint
Chest pain with no history of trauma
History
Pain

1. Ongoing and severe and crushing and substernal or same as previous pain diagnosed as MI
IV access, supplemental oxygen, cardiac monitor, ECG, aspirin, nitrates, management
of ongoing pain, admit, serum cardiac markers (eg, CKMB), CXR, anticoagulation
2. Severe or pressure or substernal or exertional or radiating to jaw, neck, shoulder, or arm
ECG, IV access, supplemental oxygen, cardiac monitor, serum cardiac markers (eg, CKMB),
CXR, nitrates, management of ongoing pain, admit
3. Tearing, severe, and radiating to back
largebore IV access, supplemental oxygen, cardiac monitor, CXR, ECG, differential
upper extremity blood pressures, aortic imaging, management of ongoing pain, admit
4. Similar to that of previous pulmonary embolus
IV access, supplemental oxygen, cardiac monitor, ABG/oximetry,
anticoagulation/pulmonary vascular imaging, ECG, CXR, admit
5. Indigestion or burning epigastric
ECG
6. Pleuritic
CXR, ECG
7. Age (male>33 years, female>40 years)
ECG

Associated Symptoms
8. Syncope or nearsyncope
ECG, cardiac monitor, Hct
9. SOB, DOE, PND, or orthopnea
ECG, ABG/oximetry, CXR
10. Significant hemoptysis
CXR, respiratory isolation, ABG/oximetry
11. Nausea/vomiting
ECG
12. Productive or chronic cough
respiratory isolation, CXR
13. Palpitations
cardiac monitor, ECG
14. Significant weight change
CXR
15. Diaphoresis
ECG

Appendix 1: Guidelines Summaries

46

Past Medical History

16. Previous MI
ECG
17. Coronary artery bypass graft/angioplasty
ECG
18. Cocaine use within last 96 hours
ECG
19. Previous positive cardiac diagnostic studies
ECG
20. Cardiac medications
serum drug levels, ECG
21. Diuretics
potassium level, magnesium level
22. IV drug abuse
ABG/oximetry, CXR, ECG
23. major risk factors for coronary artery disease (see Appendix A)
ECG
24. major risk factors for pulmonary embolism (see Appendix A)
ABG/oximetry, CXR, pulmonary vascular imaging, venous imaging, ECG
25. major risk factors for thoracic aortic aneurysm/dissection (see Appendix A)
CXR, aortic imaging, ECG
26. major risk factors for pericarditis/myocarditis (see Appendix A)
serum cardiac markers, CXR, echocardiography, ECG
27. major risk factors for pneumothorax (see Appendix A)
oximetry, CXR
28. major risk factors for pneumonia (see Appendix A)
ABG/oximetry, CXR

Physical Examination
Vital Signs
29. Irregular pulse
rhythm strip/cardiac monitor, ECG
30. Tachypnea (RR>24)
ABG/oximetry, CXR, ECG
31. Fever (>3 C/100. F)
CXR
32. Hypertension (>160/110)
CXR, ECG
33. Tachycardia (>100)
ABG/oximetry, ECG
34. Bradycardia (<60)
IV access, ECG

Appearance
35. Cyanosis with respiratory distress
IV access, supplemental oxygen, cardiac monitor, ABG, CXR, ECG, methemoglobin
level, pulmonary vascular imaging, intubation, admit
36. Diaphoresis
ECG, IV access, ABG/oximetry, serum cardiac markers, CXR, admit

Appendix 1: Guidelines Summaries

47

Cardiovascular
37. Significant differential upper extremity blood pressures
largebore IV access, CXR, ECG, supplemental oxygen, cardiac monitor, aortic imaging
38. New murmur
ECG, CXR, echocardiography
39. Pericardial rub
ECG, serum cardiac markers, CXR, echocardiography
40. Irregular rhythm .
rhythm strip/cardiac monitor, ECG
41. JVD
1CXR, ECG
42. S3 gallop
CXR, ECG

Pulmonary
43. Unilateral diminished breath sounds
CXR, ABG/oximetry
44. Localized dullness to percussion
CXR
45. Pleural rub
ABG/oximetry, CXR, ECG
46. Unilateral rales
supplemental oxygen, ABG/oximetry, CXR
47. Bilateral rales
IV access, supplemental oxygen, ABG/oximetry, CXR, ECG
48. Wheezing
supplemental oxygen, cardiac monitor, ABG/oximetry, FEV 1 /PEF, CXR, ECG,
bronchodilators

Extremities
49. Signs of DVT: leg swelling, pain, tenderness, warmth, or erythema
ABG/oximetry, CXR, pulmonary vascular imaging, venous imaging, ECG
50. Bilateral edema
CXR, ECG

Diagnostic Testing
51. cardiac monitor/ECG: new dysrhythmia
cardiac monitor, IV access, supplemental oxygen, potassium level, magnesium level, serum
cardiac markers, ECG, antidysrhythmic therapy
52. ECG: new injury
IV access, supplemental oxygen, cardiac monitor, assess for thrombolytic therapy (see
Appendix B) or other reperfusion techniques, anticoagulation, aspirin, nitrates,
management of ongoing pain, admit, serial serum cardiac markers, CXR, cardiac imaging,
serial ECGs, magnesium therapy if not given thrombolytics, bblockers
53. ECG: new ischemic changes
IV access, supplemental oxygen, cardiac monitor, anticoagulation, aspirin, nitrates,
management of ongoing pain, admit, comparison with previous ECG if available, serial
serum cardiac markers, CXR, serial ECGs, bblockers
54. ECG: nondiagnostic ECG
comparison with previous ECG if available, IV access, supplemental oxygen, cardiac
monitor, serial serum cardiac markers, CXR, serial ECGs, nitrates, management of ongoing
pain
55. CXR: acute pulmonary edema

Appendix 1: Guidelines Summaries

48

IV access, supplemental oxygen, cardiac monitor, ABG/oximetry, ECG,

preload/afterload reduction (eg, diuretics/nitrates), serum cardiac markers, management of
ongoing pain, inotropic support, admit
56. CXR: wide mediastinum (new)
largebore IV access, aortic imaging, differential upper extremity blood pressure,
supplemental oxygen, cardiac monitor, blood type and crossmatch, surgical consult
57. CXR: large pneumothorax
re-expansion, IV access, supplemental oxygen, ABG/oximetry, admit
58. CXR: small pneumothorax
IV access, ABG/oximetry, serial CXRs, re-expansion, consult
59. CXR: new infiltrate
respiratory isolation, supplemental oxygen, ABG/oximetry, appropriate cultures, sputum for
Gram stain, antibiotics, admit
60. CXR: new mass
ABG/oximetry
61. CXR: new effusion
ABG/oximetry, thoracentesis
62. ABG: new significant Aa gradient or new significant reduction in PO 2
Supplemental oxygen, CXR, IV access, cardiac monitor, pulmonary vascular imaging, ECG,
admit
63. VQ scan: high probability of pulmonary embolus
IV access, supplemental oxygen, cardiac monitor, ABG/oximetry, anticoagulation,
admit, assess for thrombolytic therapy (see Appendix B), ECG
64. VQ scan: intermediate, low, and normal in presence of high clinical suspicion
IV access, supplemental oxygen, cardiac monitor, venous imaging, pulmonary angiography,
anticoagulation, admit

Assessment
65. Unstable angina: newonset exertional
ECG, aspirin, IV access, supplemental oxygen, cardiac monitor, nitrates, consult/admit
66. Unstable angina: ongoing or recurrent ischemia
IV access, supplemental oxygen, cardiac monitor, ECG, anticoagulation, aspirin,
nitrates, management of ongoing pain, admit, serial serum cardiac markers, CXR, cardiac
imaging, serial ECGs, b blockers
67. High clinical suspicion of MI with nondiagnostic ECG
IV access, supplemental oxygen, cardiac monitor, anticoagulation, aspirin, nitrates,
management of ongoing pain, admit, serial serum cardiac markers, .CXR, cardiac imaging,
serial ECGs, magnesium therapy, bblockers
68. High clinical suspicion of MI with bundle branch block
IV access, supplemental oxygen, cardiac monitor, assessment for thrombolytic therapy
(see Appendix B) or other reperfusion techniques, anticoagulation, aspirin, nitrates,
management of ongoing pain, admit, serial serum cardiac markers, CXR, cardiac imaging,
serial ECGs, magnesium therapy if not given thrombolytics, bblockers
69. Low clinical suspicion of MI with nondiagnostic ECG .
IV access, supplemental oxygen, cardiac monitor, serial serum cardiac markers, CXR, cardiac
imaging, serial ECGs, anticoagulation, bblockers, aspirin, nitrates, management of ongoing
pain, admit
70. Acute MI with diagnostic ECG
IV access, supplemental oxygen, cardiac monitor, assessment for thrombolytic therapy
(see Appendix B) or other reperfusion techniques, anticoagulation, aspirin, nitrates,
management of ongoing pain, admit, serial serum cardiac markers, CXR, cardiac imaging,
serial ECGs, magnesium therapy if not given thrombolytics , bblockers

Appendix 1: Guidelines Summaries

49

71. Aortic dissection

largebore IV access, supplemental oxygen, cardiac monitor, blood type and
crossmatch, ECG, management of blood pressure/cardiac contractility, management of
ongoing pain, immediate surgical consultation, admit, aortic imaging
72. Pericarditis/myocarditis
ECG, serum cardiac markers, CXR, echocardiography, consult/admit
73. Pneumonia
CXR, ABG/oximetry, appropriate cultures, sputum for Gram stain, antibiotics
74. Pulmonary embolus
IV access, supplemental oxygen, cardiac monitor, ABG/oximetry, CXR, ECG,
anticoagulation, admit, assessment for thrombolytic therapy (see Appendix B), venous
imaging, consultation for filter placement if history of recurrent pulmonary embolus
75. Pneumothorax
CXR, IV access, supplemental oxygen, ABG/oximetry, serial CXRs, re-expansion,
consult/admit
76. Acute pulmonary edema
IV access, supplemental oxygen, cardiac monitor, ABG/oximetry, CXR, ECG,
preload/afterload reduction (eg, diuretics/nitrates), serum cardiac markers, management of
ongoing pain, inotropic support, admit

Disposition
77. Admission
transfer care to accepting physician
78. Transfer
follow ACEP and other applicable transfer policies
79. Discharge
provide referral for followup care, provide instructions regarding treatment and
circumstances that require return to emergency department

Notes:
Abbreviations
A-a gradient=arterial to alveolar oxygen
gradient
ABG=arterial blood gas
CXR=chest x-ray
DVT=deep vein thrombosis
DOE=dyspnea on exertion
ECG=electrocardiogram
FEV 1 /PEF=forced expiratory volume 1-second /peak expiratory flow
IV=intravenous
JVD=jugular venous distension
MI=myocardial infarction
PND=paroxysmal nocturnal dyspnea
SOB=shortness of breath

Appendix 1: Guidelines Summaries

50

APPENDIX A: RISK FACTORS

Coronary Artery Disease
Family history of coronary artery disease
Men, age > 33; Women, age > 40 2
Diabetes mellitus
Hypertension
Cigarette use
Left ventricular hypertrophy
Elevated cholesterol blood levels (High LDL/Low
HDL/Cholesterol levels)
History of chronic cocaine use
Pulmonary Embolism
Prolonged immobilization
Surgery lasting >30 minutes within the last 3
months
History of DVT or pulmonary embolus
Malignancy
Pregnancy or recent pregnancy
History of pelvis or lower extremity trauma
Oral contraceptive use combined with cigarette
smoking
Congestive heart failure
Chronic obstructive pulmonary disease
Obesity
Hypercoagulability (Antithrombin III deficiency,
etc)
Thoracic Aortic Aneurysm/Dissection
Hypertension
Congenital disease of ascending aorta or aortic
valve
Inflammatory disease of aorta
Connective tissue disease
Pregnancy
Arteriosclerosis
Cigarette use

Pericarditis/Myocarditis
Infection (eg, tuberculosis, viral)
Autoimmune/systemic disease (eg, lupus
erythematosus)
Acute rheumatic fever (autoimmune)
Recent myocardial infarction or cardiac surgery
Malignancy
Radiation therapy to mediastinum
Uremia
Drugs (Procainamide, Hydralazine, INH, etc)
History of a prior episode of pericarditis
Pneumothorax
History of previous pneumothorax
Valsalva manoeuvre
Lung disease (obstructive, cancer, infection,
connective tissue disease)
Cigarette use
Pneumonia
Chronic lung disease
Altered consciousness/impaired gag reflex
Neuromuscular disease
Thoracic cage deformity
Cigarette use
Preceding viral respiratory infection
Immunodeficiency

APPENDIX B: CONTRAINDICATIONS TO THROMBOLYTIC THERAPY

Major Contraindications
Active internal bleeding
Severe uncontrollable hypertension
History of hemorrhagic CVA
Known intracranial aneurysm, AV
malformation, or neo-plasm
Intracranial or spinal surgery within the
last 6 weeks
Significant cranial or spinal trauma within
the last 6
weeks

Other Contraindications
History of poorly controlled severe hypertension
History of nonhemorrhagic CVA
Major surgery within the last 6 weeks
Significant trauma within the last 6 weeks (including
prolonged CPR)
Gastrointestinal bleeding within the last 6 weeks
Pregnancy
Diabetic proliferative retinopathy
Likelihood of left heart thrombus (eg, mitral stenosis
with atrial fibrillation)
Recent puncture of noncompressible vessel
Age >75
Acute pericarditis

*It is important for the treating physician to realize that the above list of contraindications to thrombolytic therapy has evolved primarily as a result
of historical recommendations, with almost no supporting scientific data. The decision to use thrombolytic therapy entails an assessment of
potential adverse risks versus anticipated benefits. Subgroup analysis of existing studies has demonstrated that select subsets of patients
traditionally excluded from thrombolytic therapy receive benefits from thrombolytic therapy which outweigh the risks.

Appendix 1: Guidelines Summaries

51

Summary 4
Chest pain guideline and continuous quality improvement system
for Canadian rural emergency health care facilities
Society of Rural Physicians of Canada, 1999
Jim Thompson, MD, Charlottetown, PEI; Nick Balfour, MD, Vernon, BC; Peter Butt, MD,
Saskatoon, Sask.; Graham Dodd, MD, Salmon Arm, BC; Valerie Krym, MD, Edmonton,
Alta.; Chris Loreto, MD, Timmins, Ont.; Keith MacLellan, MD, Shawville, Que.; Jock
Murray, MD, New Glasgow, NS; Tom O'Neill, MB BCh, Shawville, Que; Mark Quigg, MD,
Collingwood, Ont.; James Rourke, MD, Goderich, Ont.; Michael Shuster, MD, Banff, Alta.;
Paul Turner, MD, Perth, Ont. CJRM 1999;4(1):9-19

Guideline overview
The guideline is divided into 4 clinical phases: "prefacility," "waiting room," "triage nurse"
and "physician". These phases are shown sequentially but can overlap. The fifth phase,
"continuous quality assurance," includes a CQI mechanism to promote self-evaluation for
identifying sources of delays that might be unique to individual REHCFs. ACIS and AMI
must always be considered in any patient presenting to an emergency department with chest
discomfort. The guideline emphasizes these disorders but does not preclude other entities in
the differential diagnosis of chest pain or discomfort.
Prefacility phase
The inability of patients to recognize symptoms of AMI is a well-known cause for delay in
treatment. Health system managers can have a variable but generally positive effect on
reducing thrombolysis times, and therefore morbidity and mortality, by promoting adequate
public education.
A significant proportion of rural patients with AMI arrive at REHCFs in private vehicles
rather than by ambulance, the latter being the usual form of transport for urban patients.
Ambulance transport, even at the basic life support level has many theoretical advantages
over private transport: a supply of oxygen, transport in a semi-Fowler's position,
administration of acetylsalicylic acid (ASA) and earlier notification of the REHCF staff.
However, no research has been done to support a recommendation regarding ambulance use
in rural settings. Some research from urban settings suggests that calling an ambulance can
result in delayed arrival at the emergency department and that the risk of cardiac arrest in
patients who travel by private vehicle is very low. On the other hand, there is evidence that
the presence of trained nurses and paramedics working in the prehospital environment can
improve significantly the recognition of AMI and subsequently early thrombolysis.
Waiting room phase
Large-volume, urban emergency departments often employ a triage nurse, who interviews all
patients as soon as they enter the waiting room to determine their priority. This level of
staffing is not practical in many REHCFs, where special solutions are required for the waiting
room phase. Participation in the CQI process can heighten the awareness of all staff who
interact with patients in the waiting room or lobby.

Appendix 1: Guidelines Summaries

52

Nursing phase (triage and initial management)

The guideline reminds nursing staff about key issues in triaging patients with chest pain and
managing those with suspected ACIS and AMI.
Cummings found that a team of at least 3 professionals at the bedside was required to speed
response (A, II-3, R-III). In many REHCFs the triage nurse often has multiple roles,
including caring for inpatients and maternity patients. A single nurse might cover the
REHCF, particularly in low-volume periods. In these situations the nurse should be able to
call other nurses for assistance, since many nursing functions should be carried out
simultaneously. In many, and perhaps most, REHCFs the triage nurse will also be one of the
nurses treating a patient with AMI; hence, we placed the triage and nursing management
phases together.
Patients who might have coronary ischemia or infarction should be triaged to a very high
urgency level owing to the risk that the patient will become unstable, the need for prompt
thrombolysis, and the need to call in additional nurses, the on-call physician and the
laboratory technician.
There is insufficient evidence to recommend a preferred sequence for nursing tasks when
limited numbers of nurses are available to assist. The sequence we show is based on the
Advanced Cardiac Life Support principle of managing the airway first, then breathing
(oxygen) and then circulation (cardiac monitor and intravenous line). The importance of a
cardiac monitor and intravenous line for patients with AMI in rural hospitals has been
indirectly demonstrated by Hindle and associates. They documented a moderate rate of
complications, such as hypotension and arrhythmia, in rural patients treated with
thrombolysis. An electrocardiogram must be obtained and interpreted as soon as possible
because it is critically important for diagnosing ACIS and AMI and for determining whether
thrombolysis is indicated. Electrocardiography should be done by a registered nurse when a
laboratory technician is not immediately available (A, III, R-III). The initial ECG should be
shown or faxed to the on-call physician very early in the triage phase.
The dosage of oxygen in chest pain is controversial. There appear to be benefits from
supplemental oxygen because hypoxemia can be associated with ongoing pain, ischemia,
arrhythmia and myocardial dysfunction. However, there is also animal evidence that oxygenfree radicals can exacerbate both infarction and reperfusion injury. No direct connection has
been made between administered oxygen dose and oxygen toxicity. The recommendation of
the American Heart Association in Advanced Cardiac Life Support 1997-99 is to start oxygen
at a rate of only 4 L/min by nasal cannula. Therefore the practice of administering high-dose
oxygen (10 to 15 L/min) to patients with chest pain but no other indication cannot be
supported (D, II-2, R-III), but the administration of, at the least, low-dose oxygen can be (B,
II-2, R-III).
Early administration of ASA is as effective as thrombolysis alone: together they produce
added benefit, but how "early" ASA must be given has not been determined. For practical
reasons, and because it seems logical to initiate ASA administration as early as possible, we
recommend that ASA be given as soon as the patient is suspected of suffering an AMI, before
the physician confirms the diagnosis. The precise timing deserves further research.
Emergency department use of ASA is an established recommendation for AMI and unstable
angina (A, I, R-III).

Appendix 1: Guidelines Summaries

53

The role of sublingual nitroglycerin is controversial. Although it can relieve ischemia and
pain in patients with angina and is widely used, its use in AMI has not been well studied. In
fact until 2 decades ago nitroglycerin was contraindicated in AMI. There is some evidence of
beneficial hemodynamic effects of sublingual nitroglycerin in AMI comparable to
intravenous nitroglycerin, although this comes with a significant increase in heart rate and
decrease in systolic blood pressure. Sublingual nitroglycerin can cause profound hypotension
in patients with AMI, particularly inferior AMI, but the precise frequency of this
complication is not well documented. One prehospital study found that sublingual
nitroglycerin was associated with profound hypotensive bradycardia, hypotension or
apnea/asystole in 4 of 300 cases. Patients with inferior AMI and ST elevation in lead V4R on
a 15-lead ECG often present with hypotension. In one study, 11% of such patients had
cardiogenic shock. Sublingual nitroglycerin can cause profound hypotension in these patients,
although the precise risk has not been measured. Hindle and associates15 found that
significant hypotension was present at some point in 23.1% of patients who received
thrombolysis in rural Alberta. Although it is not known how many of those patients had
hypotension as a result of right ventricular infarction, clearly the risk is present. Therefore
sublingual nitroglycerin should be withheld, or used with caution, if right ventricular
infarction is suspected (D, II-2, R-III). Certainly it should be withheld if the patient is
hypotensive or bradycardic (E, II-1, R-III). The on-call physician might not be present when
the decision to give or withhold nitroglycerin has to be made. This creates a conundrum for
rural nurses who must manage a patient with chest pain before a physician can confirm the
diagnosis. Further research is needed to determine the true risk of complications induced by
the use of sublingual nitroglycerin in rural settings.
Physician phase (diagnosis and definitive treatment)
The guideline encourages early physician attendance and reminds physicians about key issues
in assessing patients with chest pain with a view to ruling out or diagnosing ACIS and AMI
and initiating appropriate treatment for AMI.
CAEP recommends that the decision to initiate thrombolysis should be made only on the
advice of a qualified physician. This does not preclude thrombolysis in level 1 or 2 REHCFs
if they are appropriately staffed with trained, qualified and experienced registered nurses and
paramedics, and if a clear consultation process to a physician is established (A, II-2, R-III).
The short time objectives for initiating thrombolysis are controversial and have not been
verified with appropriate outcome research in any setting, either rural or urban. Although
myocardial rescue is clearly time-dependent, the degree of improved outcome by shortening
the objective is not clear from the literature. A number of rural physicians have reported
during Internet discussions and teleconferences that these objectives seem unrealistic in
REHCFs staffed by physicians on call-back rather than in-house, or in REHCFs where there
are no physicians in the community. It has been emphasized, however, that the time frames
are objectives, not standards of care.
We therefore recommend that the objectives for door-to-drug time should be 30 minutes for
classic AMI presentations when the physician is immediately available in the REHCF (B, I,
R-III). It should be 60 minutes when the AMI presentation is not classic, complications are
present, or when the physician has to be called back to attend the patient (B, III, R-III).
Missed opportunities for coronary thrombolysis in rural facilities have not been studied, but
they undoubtedly occur and the rate could be improved in some facilities. AMI diagnosis can
be difficult when symptoms and the ECG are atypical, and atypical presentations of AMI are

Appendix 1: Guidelines Summaries

54

common. About 50% of initial ECGs are not diagnostic of AMI. The rural nurse and
physician can improve diagnosis and the rate of AMI recognition in these situations by reevaluating the patient often, obtaining serial ECGs, using the 15-lead ECG and consulting
with colleagues. When specialist physicians are not available in the rural community then
specialist consultation should be available promptly by telephone and facsimile.
The guideline reminds physicians of important concepts in the recognition and management
of AMI. Hindle and associates found that adjunctive therapies with proven benefit in the
management of AMI were used variably in REHCFs. ASA was given in 96.7%, but betablockers in only 59.8% of cases. The American Heart Association recommends nitroglycerin
intravenously with thrombolysis. There is some evidence that intravenous nitroglycerin is
beneficial in both AMI and unstable angina, although its precise role remains unclear. There
is insufficient published information to provide practice standards for rate of use of these
adjunctive therapies.
Continuous quality improvement phase
The guideline encourages REHCF nonclinical and clinical staff to identify causes of delay
soon after each case and to capture that information on a data-entry form. This approach has
proven effective in urban settings. However, urban solutions appear to be too labour intensive
for many REHCFs, so the Committee devised a shorter dataset, which includes physician
call-back and interfacility transfer. The CQI process requires that this information be entered
into the CMM software. Both nurses and physicians should fill in a data collection form
during and soon after the episode to ensure that accurate, complete information is collected.
The CMM reporting module makes it easy for local managers to print tailored reports for
prompt feedback to staff and managers.
The CMM software can be installed on a computer in each REHCF and data entered either by
the emergency nurse or health records staff. All patients who are considered for the diagnosis
of AMI in either the emergency department or in-patient wards should be entered. REHCFs
should set up regular in-services to ensure that all hospital staff remain current and
understand the guideline (A, II-2, R-III). The CMM software reports can be used during inservices to improve guideline compliance.
Implementation of the guideline and continuous quality improvement program in
Canada
The summary wall poster and CMM software are being distributed to all Canadian rural
hospitals with 24-hour emergency departments. The rural chest pain guideline was designed
to be as generic as possible. However, so many variations exist in the staffing, inventory and
layout of REHCFs and in the geography of rural communities in Canada, that some
institutions will need to modify the guideline to suit their circumstances. This can be done by
developing a chest pain care map. It can be inserted as part of the patient chart to remind staff
about optimal care options and to record actual patient care for each individual.
Further work
Obviously much research needs to be done in many areas of managing AMI and ACIS in
rural settings. A chest pain care map should be devised for REHCFs. Research should be
conducted to determine the effectiveness of our guideline and CQI program in REHCFs. The
guideline will require modification, as new evidence becomes available regarding both the

Appendix 1: Guidelines Summaries

55

effectiveness of specific interventions in the management of AMI and the organization,

function and staffing of REHCFs. Data collected with the CMM software can be pooled to
study AMI management by REHCFs over a wide area and to identify causes of delay. Since
the same nurses and physician who cover the emergency department in many REHCFs also
cover the in-patient wards, this chest pain guideline and CQI system can also be used to
manage in-patients. We did not explore special issues that might need to be considered when
using this system for in-patients. Evidence-based guidelines should be developed for the
continued management of AMI and ACIS in REHCFs after initial emergency care, including
guidelines for inpatient management and interfacility transfer.
Acknowledgements: The CAEP Rural & Small Urban Committee is grateful to Hoffman LaRoche for funding to develop and distribute the chest pain guideline and CQI software, and to
Drs. Hugh Hindle, Tim Allen and Eric Letovsky, and the members of the CAEP Standards
Committee for their assistance.

Appendix 1: Guidelines Summaries

56

Appendix 2: Summary of AHRQ Evidence Report No. 26

Evidence Report/Technology Assessment: Number 26
Evaluation of Technologies for Identifying Acute Cardiac Ischemia
in Emergency Departments
Under its Evidence-based Practice Program, the Agency for Healthcare Research and Quality
(AHRQ) is developing scientific information for other agencies and organizations on which
to base clinical guidelines, performance measures, and other quality improvement tools.
Contractor institutions review all relevant scientific literature on assigned clinical care topics
and produce evidence reports and technology assessments, conduct research on
methodologies and the effectiveness of their implementation, and participate in technical
assistance activities.

Introduction
Acute myocardial infarction (AMI) is the leading cause of death in the United States.
Investigating the causes, progression, and treatment of AMI continues to be a national
research priority. In clinical medicine, much research has focused on the early diagnosis and
treatment of acute cardiac ischemia (ACI), which includes both unstable angina pectoris
(UAP) and AMI. In 1991, the National Heart, Lung, and Blood Institute (NHLBI) of the
National Institutes of Health initiated the National Heart Attack Alert Program (NHAAP) to
study the issues related to rapid recognition and response to patients with signs and symptoms
of ACI in emergency department (ED) settings, the point at which most of these patients
enter the health care system. This ongoing effort brings together scientists, clinicians, and
NHLBI staff with a Coordinating Committee that includes representatives of 40 professional
organizations.
In 1994, the NHAAP Working Group on Evaluation of Technologies for Identifying Acute
Cardiac Ischemia in the Emergency Department was formed to assess the technologies for
diagnosing ACI and AMI in the ED. Members of the Working Group had expertise in the
areas of cardiology, emergency medicine, general internal medicine, family practice, and
nursing, as well as in the specific disciplines of meta-analysis and health services research.
The Working Group reviewed all technologies for diagnosing ACI in the ED. The
assessments of these technologies in actual use in EDs, and the nature, extent, and quality of
the evidence on which the assessments were based, are presented in the Working Group's
final 1997 report, An Evaluation of Technologies for Identifying Acute Cardiac Ischemia in
the Emergency Department.

Reporting the Evidence

In 1998, the Agency for Healthcare Research and Quality (AHRQ, formerly the Agency for
Health Care Policy and Research [AHCPR]), working as a partner for the NHLBI's NHAAP,
contracted with the New England Medical Centers Evidence-based Practice Center(EPC) to
update the 1997 NHAAP report. The EPC was charged with evaluating the evidence on these
diagnostic technologies published since October 1994.
As before, the purpose of the review was to assess the accuracy of technologies for
diagnosing ACI in the emergency department and their clinical impact when used in this

Appendix 2: Summary of AHRQ Evidence Report 26

57

setting. However, the original 1997 report did not provide quantitative estimates of the test
performance or clinical impact of the diagnostic technologies. To address this, we conducted
meta-analyses where possible in which we reexamined all the studies reviewed in the original
report, abstracted the necessary data, and combined these data with more recently published
studies. We also conducted decision and cost-effectiveness (CE) analyses to investigate the
interactions between technologies' diagnostic performances and costs, populations, and
outcomes, and to provide an evidence-based framework on which to base recommendations.
NHAAP Working Group members help frame some of the study issues but they were not
involved in the evaluation of evidence or in the writing of the report.

Methods
We conducted a systematic and comprehensive search of the English-language literature
published between 1966 and December 1998. Literature was retrieved from a computer
MEDLINE search, references cited in the 1997 Working Group report, review of references
of retrieved articles, and assistance from domain experts. Search terms included those related
to the diagnosis of ACI, AMI, and UAP in the ED and to the following technologies:
Prehospital electrocardiography (ECG).
Continuous/serial ECG.
Non-standard leads ECG.
Exercise stress ECG.
The ACI Time-Insensitive Predictive Instrument (ACI-TIPI).
The Goldman chest pain protocol.
Biochemical tests and biomarkers (e.g., creatine kinase [CK] or its subunit
[CK-MB], troponin T, etc.).
Sestamibi myocardial perfusion imaging.
Echocardiography.
Computer-based decision aids.
Inclusion Criteria
We followed the general approach for selecting studies taken by the Working Group in its
report. We considered reports if they came from work done in the ED setting; results coming
from other settings (e.g., the cardiac care unit) were used only if little or no ED-based data
were available. Data from non-ED settings were used with the understanding that they
suggest potential utility but do not directly apply to the emergency setting.
We accepted prospective and retrospective studies that evaluated one or more of the
technologies considered in this evidence report and included patients 18 years and older who
presented to the ED with symptoms suggestive of ACI. We placed no restrictions on patients'
gender or ethnicity. In general, ED testing consists of either a single test performed within the
initial 4-hour period after presentation to the ED, or repeated testing up to 14 hours after the
patient's presentation to the ED. We accepted studies with minor deviations from this
standard.
Data were abstracted according to a written protocol and were summarized in evidence
tables.
Grading of the Evidence
The evidence-grading scheme we used assesses four dimensions that are important for the
proper interpretation of the evidence:
Size of the study (weight of the evidence).
Applicability (population category and prevalence of disease).
Appendix 2: Summary of AHRQ Evidence Report 26

58

Diagnostic performance or magnitude of clinical impact.

Methodological quality (internal validity).
Applicability. We grouped the populations and settings of the studies using a four-category
scale to help interpret the results. We also collected data about the prevalence of ACI or AMI
to assist the interpretation. The four defined population categories are:
Category IStudies that included all patients with signs and symptoms suggestive of
ACI, such as chest pain, shortness of breath, jaw pain, acute pulmonary edema, and so
forth. This is the most inclusive category. Few studies met Category I criteria.
Category IIStudies that used chest pain as the inclusion criteria. Most studies
belong to this group. Category II is a subset of Category I.
Category IIIStudies that included patients with chest pain but excluded those with
clinical or ECG findings of AMI. Many studies, especially studies of stress cardiac
imaging or testing, belong to this group. Category III is a subset of Category II.
Category IVStudies in which all patients were hospitalized or which used
additional criteria that enrolled highly selected subpopulations. We also placed
retrospective studies in this category.
Test performance studies. When there were sufficient data for a technology, we used three
complementary methods of synthesizing data across several studies to report on its test
performance:
1. Summary receiver operating characteristics (SROC) analysis.
2. Separately combined sensitivity and specificity values using a random effects model.
3. The summary diagnostic odds ratios using a random effects model.
We defined a three-level methodological quality scale for test performance studies graded as
follows:
A (least bias)Such as a study that adheres to the traditionally held concepts of high
quality diagnostic evaluation, including:
Clear descriptions of the population and setting.
Clear descriptions of the reference standard, the test under investigation, and the
diagnostic criteria.
Masked interpretation of the reference test and the test under investigation.
Verification of the diagnoses in all or most of the patients with negative results.
No significant reporting errors that are likely to result in substantial bias.
B (susceptible to some bias)A study that does not meet all the criteria in category
A, but its deficiencies are unlikely to cause major bias.
C (likely to have significant bias)A study with significant design or reporting
flaws that cannot preclude major bias. This category includes studies in which
verification bias could be a major issue and studies that have significant amounts of
missing information or discrepancies in their reporting.

Clinical impact studies. In the few instances where there are sufficient data reported by
clinical impact studies, dichotomous outcomes expressed as risk ratio or continuous outcomes
were combined using a random effects model.
We defined a three-level methodological quality scale for clinical impact studies graded as
follows:
A (least biased)Such as prospective controlled trials.
B (susceptible to some bias)Such as prospective cohort studies.
C (likely to have significant bias)Other designs or studies with significant conduct
or reporting problems that could lead to large bias.

Appendix 2: Summary of AHRQ Evidence Report 26

59

Findings
General Observations
The MEDLINE literature search identified 6,667 titles, a third of which were published from
1994 onward, indicating increased research activities on this topic over the past 5 years
compared to the previous 27 years. From these abstracts, 407 full articles were retrieved for
review, 106 of which are included in the analysis.
A diverse array of technologies with varying degrees of diagnostic accuracy is available for
use in general or selected populations to diagnose ACI in the ED. About half the studies
analyzed were in population category II and about 30 percent in category III. Prevalence of
AMI across studies, even within population categories and in similar settings, varied widely
with little indication that similarly reported inclusion criteria among studies resulted in
similar levels of AMI prevalence.
Despite this, there is some indication that overall, studies that included all patients with chest
pain (population category II) have higher prevalence of AMI than either studies that included
all patients with symptoms suggestive of ACI (population category I) or studies that excluded
patients with diagnostic ECGs (population category III). In addition, though differences in
AMI prevalence among different settings are not statistically significant, there is evidence
that studies that analyzed only admitted ED patients have higher prevalence of AMI than
those that included all ED patients. Thus, these two populations may truly be different.
Specific Findings
Most studies evaluated the accuracy of the technologies; only a few evaluated the clinical
impact of routine use. To summarize:
Prehospital 12-lead ECG has moderate sensitivity (76 percent) and specificity (88
percent) for diagnosis of ACI. It has demonstrated a reduction of the mean time to
thrombolysis by 33 minutes and short-term overall mortality in randomized trials.
In the general ED setting, only ACI-TIPI has demonstrated, in a large multicenter
clinical trial, a reduction in unnecessary hospitalizations without decreasing the rate of
appropriate admission for patients with ACI.
The Goldman chest pain protocol has good sensitivity (about 90 percent) for AMI but
has not been shown to result in any differences in hospitalization rate, length of stay
or estimated costs, in the single clinical impact study performed. Its applicability to
patients with UAP has not been evaluated.
Single measurement of biomarkers at presentation to the ED has poor sensitivity for
AMI although most biomarkers have high specificity (over 90 percent). Serial
measurements can greatly increase the sensitivity for AMI while maintaining their
excellent specificity. Biomarkers cannot identify most patients with UAP.
Diagnostic technologies to evaluate ACI in selected populations, such as
echocardiography, sestamibi perfusion imaging, and stress ECG, may have very good
to excellent sensitivity; however, they have not been sufficiently studied.

Results of Decision and Cost-Effectiveness Analyses

Decision and cost-effectiveness analyses were performed for 17 technologies and 4
combinations of technologies that have been evaluated in the literature and this report. The
cost analysis is from the payers' perspective (e.g. health insurance companies); patient
outcomes are either appropriate triage or 30-day survival of patients with ACI.

Appendix 2: Summary of AHRQ Evidence Report 26

60

As not all technologies can be applied to all patients in the ED (such as stress ECG), two
different ED populations were used for the analysis:
A general population model, which includes all patients in the ED.
A subgroup model, in which high-risk patients are excluded.
Stress tests, sestamibi imaging, and serial and continuous ECG were evaluated only in the
subgroup population.
As expected, technologies with the best diagnostic accuracy for AMI and UAP have the
highest values for appropriate triage for patients with ACI. Technologies that are more
effective (greater number of patients with ACI appropriately triaged) tend to have higher total
costs, with the exception of ACI-TIPI. The biomarkers are least costly and have the lowest
values for appropriate triage. Algorithms, combination technologies, and echocardiography
are the next most effective technologies, in that order. Sestamibi imaging and exercise ECG
are more expensive than other technologies but have excellent diagnostic performance for
ACI.
Based on data using only the diagnostic performance data of technologies, the combination
technology of troponin T and echocardiography has the best CE among all technologies
applicable to the general population model. If results from clinical impact studies are
incorporated, ACI-TIPI has the best CE because of its very high triage accuracy and low cost.
The incremental CE of troponin T and echocardiography is about $7,670 per additional
appropriate triage for a patient with ACI compared with serial or combination biomarkers.
The incremental CE of the next most effective technology, the artificial neural network, is
approximately $10,560. Given the economic ramifications and the effects on the patient of a
missed ACI diagnosis, this incremental CE for troponin T and echocardiography is minimal.
Because the estimates for detection of UAP are based on sparse data, we also evaluated the
triage accuracy and cost-effectiveness of technologies for appropriate triage for patients with
AMI only. The relative CE rankings do not change compared with the rankings for patients
with ACI. There are few but important differences, however, in triage accuracy:
1. The Goldman protocol improves significantly.
2. Serial CK-MB improves slightly.
3. The combination of troponin T and echocardiography is slightly better than ACI-TIPI
(a difference of one patient with AMI appropriately triaged).
The combination of troponin T and echocardiography is the most cost-effective, followed by
the artificial neural network. The incremental CE between these two technologies is much
larger than in the general ACI model: approximately $137,000 per additional appropriately
triaged patient with AMI.
In the low-risk patient subgroup model, ACI-TIPI is again the most cost-effective technology
if data from clinical impact studies are incorporated. Sestamibi stress imaging has the best
diagnostic performance (detects 82 percent of patients with ACI), followed by sestamibi rest
scanning, and exercise ECG. The costs of exercise ECG and stress sestamibi are nearly the
same. The incremental CE between the two technologies is a mere $364 per appropriately
triaged patient, reflecting the higher effectiveness of stress sestamibi for its cost relative to
exercise ECG.
The incremental CE between stress sestamibi imaging and the next cost-effective technology,
the combination of troponin T and echocardiography, is much greater: $12,757. However,
given that stress sestamibi imaging results in the appropriate triage of 37 additional patients

Appendix 2: Summary of AHRQ Evidence Report 26

61

with ACI (per 1,000 ED patients) compared with troponin T and echocardiography, it appears
to be a very cost-effective technology.
If data from the ACI-TIPI trial are used, the incremental CE of using ACI-TIPI compared
with troponin T and echocardiography is only $1,502 per additional appropriate triage for a
patient with ACI, a truly negligible increase for improved triage accuracy.
Considering only triage accuracy for patients with AMI, the combination of troponin T and
echocardiography is the most cost-effective. Exercise ECG and stress sestamibi imaging also
have excellent triage accuracy; however, the per ED patient costs of these two technologies is
about $500 more than that of troponin T and echocardiography.

Future Research

Most studies evaluated the performance of a technology in diagnosing AMI; future

studies should also evaluate a technology's performance in diagnosing UAP.
Some technologies (e.g., echocardiography, sestamibi imaging, exercise ECG, serial
biomarkers, and new biomarkers such as P-selectin and fatty acid binding proteins)
remain under-evaluated.
To date, most studies have evaluated the application of a single technology on
patients. Research is needed to determine whether combinations of tests, such as a
panel of biomarkers, or of multiple modalities, such as ECG with serial CK-MB
measurements, perform better than the component tests alone.
Because good test performance, in isolation, does not automatically translate to
appropriate utilization or desired outcomes, clinical impact studies are needed to
evaluate the clinical outcomes of the actual use of the test.
The prevalence of ACI among the studies varies widely and may be explained only
partially by differences in patient populations. The wide variation of prevalence has
an unknown effect on test performance and interpretation of the results, and may
indicate incomplete reporting of study biases. We need to understand the reason for
the heterogeneity of the prevalence among studies with seemingly similar patient
populations.
The methodological quality and the reporting of the diagnostic performance studies
on this topic varies widely and could be improved substantially.

Availability of Full Report

The full evidence report from which this summary was taken was prepared for the Agency
for Healthcare Research and Quality by the New England Medical Center under contract No.
290-97-0019. Printed copies may be obtained free of charge from the AHRQ Publications
Clearinghouse by calling 1-800-358-9295. Requesters should ask for Evidence
Report/Technology Assessment No. 26, Evaluation of Technologies for Identifying Acute
Cardiac Ischemia in Emergency Departments (AHRQ Publication No. 01-E006).
The Evidence Report can also be downloaded as a zipped file online at:
http://www.ahrq.gov/clinic/evrptfiles.htm.
AHRQ Publication No. 00-E031
Current as of September 2000

Appendix 2: Summary of AHRQ Evidence Report 26

62

Appendix 3 Abstracts of all primary publications which were

critically appraised
Prehospital 12-Lead Electrocardiography
No new data.

Continuous/Serial 12-Lead ECG

No new data specifically on this area

Nonstandard Lead ECG

Aufderheide TP, Xue Q, Dhala AA, Reddy S, Kuhn EM. The added diagnostic value of
automated QT-dispersion measurements and automated ST-segment deviations in the
electrocardiographic diagnosis of acute cardiac ischemia. Journal of
Electrocardiology 2000;33(4):329-39.
The purpose of this study was to determine the added value of automated QT dispersion and ST-segment
measurements to physician interpretation of 12-lead electrocardiograms (ECGs) in patients with chest pain. To
date, poor reproducibility of manual measurements and lack of shown added value have limited the clinical use
of QT dispersion. Twelve-lead ECGs (n = 1,161) from the Milwaukee Prehospital Chest Pain Database were
independently classified by 2 physicians into 3 groups (acute myocardial infarction (AMI), acute cardiac
ischemia (ACI), or nonischemic), and their consensus was obtained. QT-end and QT-peak dispersions were
measured by a computerized system. The computer also identified ST-segment deviations. Sensitivity,
specificity, and positive predictive values (PPVs) and negative predictive values (NPV) for AMI and ACI were
evaluated independently and in combinations. For AMI, physicians' consensus classification was remarkably
good (sensitivity, 48%, specificity, 99%). Independent classification by QT-end and QT-peak dispersions or ST
deviations was not superior to the physicians' consensus. Optimal classification occurred by combining
automated QT-end dispersion and ST deviations with physicians' consensus. This combination increased
sensitivity for the diagnoses of AMI by 35% (65% vs 48%, P < .001) and ACI by 55% (62% vs 40%, P < .001)
compared with physicians' consensus, while maintaining comparable specificity. This study supports a potential
clinical role for automated QT dispersion when combined with other diagnostic methods for detecting AMI and
ACI.

Exercise Stress Testing

Buchsbaum M, Marshall E, Levine B, et al. Emergency department evaluation of chest pain
using exercise stress echocardiography. Academic Emergency Medicine
2001;8(2):196-9.
OBJECTIVE: Patients with a low risk of coronary artery disease (CAD) presenting to the emergency
department (ED) with chest pain pose a diagnostic dilemma because a small percentage will suffer an acute
myocardial infarction (MI) and sudden death. The authors conducted this study to determine whether exercise
stress echocardiography (ESE) could be used to further support the safe discharge of these low-risk patients.
METHODS: A convenience sample of patients > or =30 years of age without a prior cardiac history who
presented to an academic community hospital with chest pain, normal initial creatine kinase, and
electrocardiography without ischemic changes underwent ESE within 6 +/- 1.7 hours (mean +/- SD). Abnormal
ESE was defined as regional wall motion abnormality at rest or after exercise. The ED disposition and threeand six-month follow-up for cardiac events were recorded. This was a prospective observational cohort study.
RESULTS: Of a total of 149 eligible patients, 145 completed the study. The mean age (+/-SD) was 47 +/- 9
years; 56% were male. No adverse events were noted during ESE. Seven patients (5%) had abnormal ESE (2
with rest wall motion abnormalities and 5 with exercise-induced wall motion abnormalities). Five of the seven
underwent cardiac catheterization; three had CAD. All patients received telephone follow-up at three months
and six months. Of the 138 patients with a normal ESE, all were free of cardiac events at three months. One

Appendix 3: Abstracts of all critically appraised primary publications

63

patient had a non-Q-wave MI at six months (negative predictive value = 99.3%, 95% CI = 97.8% to 100%).
CONCLUSIONS: Exercise stress echocardiography can be used to evaluate low-risk chest pain patients in the
ED. Patients with a normal ESE may be considered for discharge with minimal risk of sequelae.

Geleijnse ML, Elhendy A, Kasprzak JD, et al. Safety and prognostic value of early
dobutamine-atropine stress echocardiography in patients with spontaneous chest pain
and a non-diagnostic electrocardiogram. European Heart Journal 2000;21(5):397406.
AIMS: To risk stratify and shorten hospital stay in patients with spontaneous (resting) chest pain and a nondiagnostic electrocardiogram (ECG). METHODS AND RESULTS: The study comprised 102 patients (mean
age 58+/-12 years, 67 men) with spontaneous chest pain and a non-diagnostic ECG. Forty-three patients had
suspected coronary artery disease and 59 had known (but of unknown actual significance) coronary artery
disease. All patients underwent serial creatine kinase enzyme measurements, continuous ECG monitoring for at
least 12 h and early dobutamine-atropine stress echocardiography in patients with negative creatine kinase
enzymes and normal findings at ECG monitoring. Dobutamine-atropine stress echocardiography was considered
positive in patients with new or worsening wall thickening abnormalities. Patients with negative dobutamineatropine stress echocardiography were discharged after the test. In-hospital and 6 month follow-up events noted
were cardiac death, non-fatal myocardial infarction, unstable angina, and coronary artery bypass surgery or
angioplasty. Thirteen patients had evidence of evolving myocardial infarction by elevated creatine kinase
enzymes, or unstable angina by ECG monitoring. In the remaining 89 patients, dobutamine-atropine stress
echocardiography was performed after a median observation period of 31 h (range 12-68 h). During
dobutamine-atropine stress echocardiography no serious complications (death, non-fatal myocardial infarction,
sustained ventricular tachycardia or ventricular fibrillation) occurred. Dobutamine-atropine stress
echocardiography results were of poor quality in three, non-diagnostic in six, negative in 44 and positive in 36
patients. In the 80 patients with diagnostic dobutamine-atropine stress echocardiography, variables associated
with in-hospital events (n=7) were history of exertional angina (P<0. 005), chest pain score (P<0.005), stressinduced angina (P<0.001) and positive dobutamine-atropine stress echocardiography (P<0.005). Variables
associated with follow-up events (n=11) were history of exertional angina (P<0.05), chest pain score (P<0.001),
stress-induced angina (P<0.01) and positive dobutamine-atropine stress echocardiography (P<0.01). At
multivariate analysis the only significant predictor of events was positive dobutamine-atropine stress
echocardiography (P<0.01). CONCLUSION: Early dobutamine-atropine stress echocardiography may safely
distinguish between low- and high-risk subsets for subsequent cardiac events in patients with spontaneous chest
pain and a non-diagnostic ECG. Copyright 2000 The European Society of Cardiology.

Krasuski RA, Hartley LH, Lee TH, Polanczyk CA, Fleischmann KE. Weekend and holiday
exercise testing in patients with chest pain. Journal of General Internal Medicine
1999;14(1):10-4.
OBJECTIVE: To determine the outcome, safety, and possible cost savings of patients undergoing weekend or
holiday exercise treadmill testing. DESIGN: Medical records of all 195 patients scheduled for weekend and
holiday exercise testing were reviewed, and 77.9% of patients were contacted by telephone to ascertain medical
outcomes and need for further emergency department or inpatient care. Costs were calculated from estimates of
days of hospitalization saved and incremental costs incurred in conjunction with weekend or holiday testing.
SETTING: Urban tertiary care academic medical center. PATIENTS: A total of 195 patients were scheduled for
testing, and 181 tests were performed. Over three quarters (75.1%) of patients underwent testing for assessment
of chest pain. Other indications included risk stratification after myocardial infarction or coronary angioplasty or
prior to noncardiac surgery, or evaluation for arrhythmias, dyspnea, or syncope. MEASUREMENTS AND
MAIN RESULTS: Outcomes included results and complications of testing, hospital course after testing,
subsequent emergency department visits and readmissions, myocardial infarction, need for cardiac
catheterization or revascularization, and mortality. No complications were noted during testing. In 136 patients
tested for the indication of chest pain, 90 (66.2%) had negative tests, 39 (28. 7%) were intermediate, and 6
(4.4%) were positive for ischemia. Same day discharge occurred in 115 (84.6%) of the patients, saving an
estimated 185 days of hospitalization ($316.83 per patient tested). Event rates over the 6 months following
discharge were low. CONCLUSIONS: Weekend and holiday exercise testing is a safe and effective means of
risk stratification prior to hospital discharge for patients with chest pain. It also reduces length of stay and is cost
saving.

Appendix 3: Abstracts of all critically appraised primary publications

64

Lewis WR, Amsterdam EA, Turnipseed S, Kirk JD. Immediate exercise testing of low risk
patients with known coronary artery disease presenting to the emergency department
with chest pain. Journal of the American College of Cardiology 1999;33(7):1843-7.
OBJECTIVES: The purpose of this study was to demonstrate the safety and utility of immediate exercise
treadmill testing (IETT) of low risk patients presenting to the emergency department with known coronary
artery disease (CAD). BACKGROUND: More than 70% of the two million patients admitted to U.S. hospitals
annually for suspected acute myocardial infarction (AMI) are found not to have had a cardiac event. We have
previously demonstrated the safety and efficacy of IETT of selected low risk patients without known CAD
presenting to the emergency department with chest pain. This study extends this approach to selected patients
with a history of CAD. METHODS: One hundred patients evaluated by the chest pain emergency room to rule
out AMI underwent IETT using a modified Bruce protocol upon admission to the hospital (median time <1 h).
RESULTS: Twenty-three patients (23%) had positive exercise electrocardiograms (ExECGs); an uncomplicated
non-Q wave AMI was diagnosed in two patients. Thirty-eight patients (38%) had negative ExECGs and 39
patients (39%) had nondiagnostic ExECGs. Of these 100 patients, 64 were discharged immediately after IETT,
19 were discharged in less than 24 h after negative serial cardiac enzymes and stable electrocardiograms and 17
were discharged after further evaluation and treatment. There were no complications from exercise testing and
no late deaths or AMI during six-month follow-up. CONCLUSIONS: Immediate exercise treadmill testing of
low risk patients with chest pain and known CAD is effective in further stratifying this group into patients who
can be safely discharged and those who require hospital admission.

Sarullo FM, Di Pasquale P, Orlando G, et al. Utility and safety of immediate exercise testing
of low-risk patients admitted to the hospital with acute chest pain. International
Journal of Cardiology 2000;75(2-3):239-43.
It is common practice to hospitalize patients with chest pain for a period of observation and to perform further
diagnostic evaluation such as exercise treadmill testing (ETT) once acute myocardial infarction (AMI) has been
excluded. This study evaluates the safety and efficacy of immediate ETT for patients admitted to the hospital
with acute chest pain. One hundred and ninety non-consecutive low-risk patients admitted to the hospital from
emergency department with acute chest pain underwent ETT using Bruce protocol immediately on admission to
the hospital (median time 165+30 min). Fifty-seven (30%) patients had positive exercise electrocardiograms, 44
(77.2%) of whom had significant coronary narrowing by angiography. An uncomplicated anterior non-Q-wave
AMI was diagnosed in one patient. One hundred and eleven (58.4%) patients had negative and 22 (11.6%)
patients had non-diagnostic exercise electrocardiograms. Of these 133 patients, 86 (64.7%) were discharged
immediately after ETT, 19 (14.3%) were discharged within 24 h, and 28 (21%) were discharged after 24 h of
observation. There were no complications from ETT. During the 17+/-6 months follow-up no patients died, and
only eight (7.2%) patients with negative ETT experienced a major cardiac event (one AMI and seven angina). In
conclusion, our results suggest that immediate ETT of low-risk patients with chest pain who are at sufficient risk
to be designated for hospital admission, is effective in further stratifying this group into those who can be safety
discharged immediately and those who require hospitalization.

Biomarkers
Creatine Kinase (CK), Single and Serial Measurements
Bock JL, Brogan GX, Jr., McCuskey CF, Thode HC, Jr., Hollander JE, Gunther T.
Evaluation of CK-MB isoform analysis for early diagnosis of myocardial infarction.
Journal of Emergency Medicine 1999;17(1):75-9.
Measurement of CK-MB and its isoforms by high-voltage electrophoresis has been proposed as a sensitive test
for early detection of myocardial infarction (MI). We performed a prospective study of this test in 231 patients
presenting to the Emergency Department with symptoms consistent with ischemic chest pain. Blood specimens
were obtained at 0, 1, and 3 h following presentation, and plasma was immediately frozen and analyzed within 1
week by high-voltage electrophoresis for total CK-MB and isoforms. The test was considered positive whenever
total CK-MB was elevated (>6 U/L) or the cardiac isoform MB2 was relatively increased (MB2 > 2 U/L and
MB2/MB1 > 1.7). This test had a sensitivity of 68% overall and 55% for specimens collected within 3 h of
symptom onset. It was positive within 3 h of presentation in 36/39 (92%) of patients with confirmed MI.
Specificity was 92% overall and did not vary with time after symptoms. The CK-MB alone, at the cutoff of 6
U/L, had lower sensitivity overall (56%; p = 0.01) and within 3 h of onset (39%; p = 0.03), and higher

Appendix 3: Abstracts of all critically appraised primary publications

65

specificity overall (98%; p < 0.001). Lowering the cutoff for CK-MB alone to match the sensitivity of the
isoform test caused a greater loss of specificity. It is concluded that analysis of CK-MB by high-voltage
electrophoresis is an effective method for rapid diagnosis of MI, with the isoform analysis enhancing early
sensitivity.

Creatine Kinase Subunit (CK-MB), Single and Serial Measurements

Bholasingh R, de Winter RJ, Fischer JC, Koster RW, Peters RJ, Sanders GT. Safe discharge
from the cardiac emergency room with a rapid rule-out myocardial infarction protocol
using serial CK-MB(mass). Heart 2001;85(2):143-8.
OBJECTIVE: To determine whether a new protocol, using a rapid and sensitive CK-MB(mass) assay and serial
sampling, can rule out myocardial infarction in patients with chest pain and decrease their length of stay in the
cardiac emergency room without increasing risk. DESIGN: The combined incidence of cardiac death and acute
myocardial infarction at 30 days, six months, and 24 months of follow up were compared between patients
discharged home from the cardiac emergency room after ruling out myocardial infarction with a CKMB(activity) assay in 1994 and those discharged home after a rapid CK-MB(mass) assay in 1996. SETTING:
Cardiac emergency room of a large university hospital. PATIENTS: In 1994 and 1996, 230 and 423 chest pain
patients, respectively, were discharged home from the cardiac emergency room with a normal CK-MB and an
uneventful observation period. RESULTS: The median length of stay in the cardiac emergency room was
significantly reduced, from 16.0 hours in 1994 to 9.0 hours in 1996 (p < 0.0001). Mean event rates in patients
from the 1994 and 1996 cohorts, respectively, were 0.9% (95% confidence interval (CI) -0.3% to 2.1%) v 0.7%
(95% CI -0.1% to 1. 5%) at 30 days, 3.0% (95% CI 0.8% to 5.2%) v 2.8% (95% CI 1.2% to 4. 4%) at six
months, and 7.0% (95% CI 3.7% to 10.3%) v 5.7% (95% CI 3. 5% to 7.9%) at 24 months. Kaplan-Meier
survival analysis showed no difference in mean event-free survival at 30 days, six months, and 24 months of
follow up. CONCLUSIONS: Using a rule-out myocardial infarction protocol with a rapid and sensitive CKMB(mass) assay and serial sampling, the length of stay of patients with chest pain in the cardiac emergency
room can be reduced without compromising safety.

Fesmire FM. A rapid protocol to identify and exclude acute myocardial infarction:
continuous 12-lead ECG monitoring with 2-hour delta CK-MB. American Journal of
Emergency Medicine 2000;18(6):698-702.
A prospective observational study was performed in 706 chest pain patients who underwent our chest pain
evaluation protocol which consists of continuous 12-lead ST-segment monitoring with automated serial ECG
(SECG) and a 2-hour delta (delta) CK-MB level determination before ED physician making final disposition
decision to determine the incremental value of our 2-hour protocol for identifying myocardial infarction (MI) as
compared with the initial ECG in combination with a baseline CK-MB. The initial ECG was obtained on
presentation and considered positive if it revealed injury or ischemia. SECGs were obtained at least every 10
minutes and considered positive if it revealed new injury or ischemia. The baseline CK-MB value was
considered positive if it was > or =12 ng/mL and index > or =4%. ACK-MB was defined as a difference
between the 2 hour and baseline CK-MB and was considered positive if the value was > or =+1.5 ng/mL. MI
was defined as acute myocardial infarction (AMI) or recent AMI (ie, AMI patients presenting on falling curve
of CK-MB). The incremental value of the 2 hour protocol (ie, SECG in conjunction with deltaCK-MB) was
more sensitive for identification of MI than the baseline protocol (ie, initial ECG in conjunction with the
baseline CK-MB) (94.0% versus 55.4%; P < .0001) and reliably both identified (+LR = 14.6) and excluded MI
(-LR = 0.06). SECG monitoring in conjunction with the 2 hour deltaCK-MB allows for early identification and
exclusion of MI, and can assist the ED physician in making appropriate treatment and disposition decisions.

Fesmire FM. Delta CK-MB outperforms delta troponin I at 2 hours during the ED rule out of
acute myocardial infarction. American Journal of Emergency Medicine 2000; 18(1):18.
It has been shown that a rise in creatine kinase MB bank (CK-MB) of > or = + 1.6 ng/mL in 2 hours is more
sensitive and equally specific for detection of acute myocardial infarction (AMI) as compared with a 2-hour
CK-MB > or = 6 ng/mL during the emergency department (ED) evaluation of chest pain. Because cardiac
specific troponin I (cTnI) is thought to have similar early release kinetics as compared with CK-MB mass, we
undertook a retrospective cohort study in 578 chest pain patients whose baseline CK-MB and cTnI was less than
two times the hospital's upper limits of normal and who underwent a 2-hour CK-MB and cTnI to compare
sensitivities and specificities of the 2-hour delta CK-MB (deltaCK-MB) and delta cTnI (delta cTnI) for AMI and

Appendix 3: Abstracts of all critically appraised primary publications

66

30-day Adverse Outcome (AO). Thirty day AO was defined as AMI, life-threatening complication, death, or
percutaneous transluminal coronary angioplasty (PTCA)/coronary artery bypass graft (CABG) within 30 days of
ED presentation. Optimum delta values were determined by choosing the smallest cutoff value greater than the
assay precision where the deltaCK-MB and delta cTnI had a positive likelihood ratio for 30-day AO of > or =
15. A deltaCK-MB > or = +1.5 ng/mL was more sensitive than a deltaTnI > or = +0.2 ng/mL for AMI (87.7%
versus 61.4%; P < .0005) and 30-day AO (56.7% versus 42.3%; P < .005). There were no differences in
specificities for AMI and 30-day AO. Combining the two tests (MBdelta > or = +1.5 ng/mL and/or a deltaTnI >
or = +0.2 ng/mL) resulted in an incremental increase in sensitivity of 89.5% for AMI and 61.9% for AO (P <
.005). Patients with either a rise in CK-MB of > or = +1.5 ng/mL or rise in cTnI of > or = +0.2 ng/mL in 2 hours
should receive consideration for aggressive antiischemic therapy and further diagnostic testing before making an
exclusionary diagnosis of nonischemic chest pain.

Green GB, Dehlinger E, McGrievey TS, et al. CK-MB isoforms for early risk stratification of
emergency department patients. Clinica Chimica Acta 2000;300(1-2):57-73.
The potential clinical utility of single sample CK-MB isoforms measurement for early risk stratification of
Emergency Department (ED) patients with possible myocardial ischemia was evaluated among 405 patients
presenting to two urban EDs. Clinical and serologic data were prospectively collected and the occurrence of
adverse events (AEs) and myocardial infarction (MI) during the 14-day outcome period was recorded and
utilized to calculate and compare relative risks (RR) and predictive values of isoforms and CK-MB alone.
Among the 405 patients, 67 accrued 105 AEs. Both isoforms and CK-MB alone were predictive of AEs with RR
of 3.32 (2.09, 5.27) and 6.28 (4.64, 8.52), respectively. Isoforms had higher sensitivity for AEs compared to
CK-MB (65.7% [54.3, 77.0] vs. 14.9% [6.4, 23.5]; p<0. 01) but lower specificity (69.2% [64.3, 74.2] vs. 99.7%
[99.1,100. 0]; p<0.01). Isoforms' superior sensitivity allowed identification of many high risk patients missed by
CK-MB alone. Further, for the prediction of MI, isoforms had superior diagnostic sensitivity and equivalent
specificity. This investigation supports the emergency department use of early, single sample CK-MB isoform
testing.

Herren KR, Mackway-Jones K, Richards CR, Seneviratne CJ, France MW, Cotter L. Is it
possible to exclude a diagnosis of myocardial damage within six hours of admission
to an emergency department? Diagnostic cohort study. BMJ 2001;323(7309):18.
Objective: To assess the clinical efficacy and accuracy of an emergency department based six hour rule-out
protocol for myocardial damage. Design: Diagnostic cohort study. Setting: Emergency department of an inner
city university hospital. Participants: 383 consecutive patients aged over 25 years with chest pain of less than 12
hours' duration who were at low to moderate risk of acute myocardial infarction. Intervention: Serial
measurements of creatine kinase MB mass and continuous ST segment monitoring for six hours with 12 leads.
Main outcome measure: Performance of the diagnostic test against a gold standard consisting of either a 48 hour
measurement of troponin T concentration or screening for myocardial infarction according to the World Health
Organization's criteria. Results: Outcome of the gold standard test was available for 292 patients. On the
diagnostic test for the protocol, 53 patients had positive results and 239 patients had negative results. There were
18 false positive results and one false negative result. Sensitivity was 97.2% (95% confidence interval 95.0% to
99.0%), specificity 93.0% (90.0% to 96.0%), the negative predictive value 99.6%, and the positive predictive
value 66.0%. The positive likelihood ratio was 13.9 and the negative likelihood ratio 0.03. Conclusions: The six
hour rule-out protocol for myocardial infarction is accurate and efficacious. It can be used in patients presenting
to emergency departments with chest pain indicating a low to moderate risk of myocardial infarction.

Peacock WF, Emerman CL, McErlean ES, et al. Normal CK, elevated MB predicts
complications in acute coronary syndromes. Journal of Emergency Medicine
2001;20(4):385-90.
The implications of an elevated Creatine kinase (CK)-MB isoenzyme (MB) in suspected acute coronary
syndromes, with a normal total CK, is not well established. Despite many guidelines on managing patients with
acute coronary ischemia, none indicates strategies for patients with elevated MB and with a normal CK. The
outcome consequence of this result is not firmly established. Our objective was to prospectively evaluate
outcomes in patients with suspected acute coronary syndromes, normal initial total CK, and increased MB. All
Emergency Department patients with suspected acute coronary syndromes and creatinine < 2.0 mg/dL were
eligible for study entry. Serial CK and MB fractions were measured on arrival in the Emergency Department,
then 8 and 16 h postpresentation. A composite outcome of death, Q-wave myocardial infarction, or
revascularization was defined at the index visit and 6 months later. Outcomes were determined by blinded
record review and by telephone contact. In the 698 patients entered, the acute composite outcome rate was 25%

Appendix 3: Abstracts of all critically appraised primary publications

67

(175) and 6.3% (44) at 6 months. Acute and 6 month adverse outcome rates were statistically the same for all
patients with an elevated MB fraction, regardless of the total CK level. An elevated MB conferred a higher
event rate than did a normal MB. We conclude that the adverse event rate for patients with suspected acute
coronary syndromes and an elevated MB is the same whether or not the total CK is elevated. These patients
should be considered as having had an acute coronary syndrome.

Troponin T and Troponin I

Johnson PA, Goldman L, Sacks DB, et al. Cardiac troponin T as a marker for myocardial
ischemia in patients seen at the emergency department for acute chest pain. American
Heart Journal 1999;137(6):1137-44.
BACKGROUND: Identification of patients with acute chest pain at high risk for cardiovascular complications
is a common and difficult challenge for clinicians and must be based initially on data from the history, physical
examination, electrocardiogram, and chest radiograph. Some data suggest that elevations in cardiac troponin T
(cTnT) may be useful for detection of less severe degrees of myocardial injury that may occur in some patients
with unstable angina. Therefore we designed a prospective follow-up study to assess the diagnostic performance
and prognostic value of cTnT in a population of patients presenting to the emergency department with acute
chest pain. METHODS: The patient population included all 1477 admitted patients aged 30 years or more who
presented to the emergency department of an urban teaching hospital from October 1992, through February
1994, with a chief symptom of acute chest pain not explained by trauma or chest radiograph abnormalities. The
1303 patients (88%) who had 2 or more measurements of cTnT during the first 24 hours after presentation
comprised the final study population. Sensitivity, specificity, positive predictive value, negative predictive
value, and receiver operator characteristics curve (ROC) were determined for cTnT and creatine kinase-MB
(CK-MB) (measured using activity and mass assays) data from the first 24 hours after admission for the
outcomes of acute myocardial infarction (AMI) and major cardiac events during the first 72 hours of
hospitalization. RESULTS: The sensitivity and specificity of cTnT (threshold of 0.1 ng/mL) for detecting AMI
during the first 24 hours after presentation were 99% and 86%, respectively. The CK-MB activity and mass
assays had diagnostic performance for detecting AMI similar to cTnT. Among patients who did not meet study
criteria for AMI, cTnT was elevated during the first 24 hours in 31% of patients who had major complications,
compared with a 17% rate for the CK-MB activity assay and a 3% rate for the CK-MB mass assay. In these
patients, the cTnT assay had superior diagnostic performance compared with the CK-MB mass assay as a
marker for cardiac complications as assessed with ROC analysis (P <.0004). CONCLUSIONS: In a
heterogeneous population of patients seen in the emergency department with acute chest pain, cTnT was similar
to CK-MB (activity and mass assays) for detection of AMI and superior to the CK-MB mass assay as a marker
for major cardiac events early in the hospital course among those who were ruled out for an AMI. Further study
is required to determine how this assay can be used to provide more appropriate, cost-effective care.

Peacock WI, Emerman CL, McErlean ES, et al. Prediction of short- and long-term outcomes
by troponin T levels in low-risk patients evaluated for acute coronary syndromes.
Annals of Emergency Medicine 2000;35(3):213-20.
STUDY OBJECTIVE: Recent reports suggest a short series of cardiac troponin (cTnT) testing effectively
identifies patients at risk for cardiac events. However, there are few studies validating this strategy. The purpose
of this study was to determine the ability of cTnT levels to predict short- and long-term outcomes in low-risk
patients with suspected acute coronary syndromes. METHODS: This prospective longitudinal study was
conducted in a 20-bed emergency department observation unit. Patients at low risk for acute coronary ischemia,
with a normal creatine kinase-isoenzyme subunit MB (CKMB) index, were admitted to an observation unit for
chest pain evaluation. Serum cTnT levels were measured at baseline and at 4, 8, and 16 hours after admission.
The main outcome measures were adverse cardiac events (death, acute myocardial infarction, unstable angina,
revascularization) during the index visit and within 6 months after discharge. Using manufacturer's
recommendations, the cTnT level was considered abnormal if it exceeded 0.2 microg/L. RESULTS: Two
hundred sixty-six patients were evaluated. Twenty-one (7.9%) had an adverse event during their index
hospitalization. Troponin testing identified only 2 (9.5%) of these patients. Twenty (7.5%) had a cardiac event
within 6 months; none were identified by cTnT testing. The sensitivity and specificity were 9.5% and 99.2%,
respectively, at the index visit, and 0% and 98.4% at 6 months. The positive and negative predictive values were
50% and 93%, respectively, at the index visit; and 0% and 92% at 6 months. CONCLUSION: Determination of
troponin T levels has a low sensitivity and high specificity for predicting outcomes in low-risk patients
evaluated for suspected acute coronary syndromes. This study does not support a strategy of relying solely on
troponin testing for disposition decisions.

Appendix 3: Abstracts of all critically appraised primary publications

68

Van Lente F, McErlean ES, DeLuca SA, Peacock WF, Rao JS, Nissen SE. Ability of
troponins to predict adverse outcomes in patients with renal insufficiency and
suspected acute coronary syndromes: a case-matched study. Journal of the American
College of Cardiology 1999;33(2):471-8.
OBJECTIVES: The purpose of this study was to investigate the utility of cardiac troponin T and troponin I for
predicting outcomes in patients presenting with suspected acute coronary syndromes and renal insufficiency
relative to that observed in similar patients without renal disease. BACKGROUND: Cardiac troponin T and
troponin I have shown promise as tools for risk stratification of patients with acute coronary syndromes.
However, there is uncertainty regarding their cardiac specificity and utility in patients with renal disease.
METHODS: We measured troponin T, troponin I and creatine kinase MB in 51 patients presenting with
suspected acute coronary syndromes and renal insufficiency and in 102 patients without evidence of renal
disease matched for the same peak troponin T or I value, selected from a larger patient cohort. Blood samples
were obtained at presentation to an emergency room 4 hours, 8 hours and 16 hours later. The ability of
biochemical markers to predict adverse outcomes in both groups including infarction, recurrent ischemia, bypass
surgery, heart failure, stroke, death or positive angiography/angioplasty during hospitalization and at six months
was assessed by receiver-operator curve analysis. The performance of both troponins was compared between
groups. RESULTS: Thirty-five percent of patients in the renal group and 45% of patients in the nonrenal group
experienced an adverse initial outcome; over 50% of patients in all groups had experienced an adverse outcome
by 6 months, but these differences were not significant. The area under the curve (AUC) for the ROC curve for
troponin T as predictor of initial outcomes was significantly lower in the renal group than in the nonrenal group:
0.56+/-0.07 and 0.75+/-0.07, respectively. The area under the curve was also significantly lower in the renal
group compared with the nonrenal group for troponin T as predictor of six month outcomes: 0.59+/-0.07 and
0.74+/-0.07, respectively. The area under the curve was also significantly lower in the renal group compared to
the nonrenal group for troponin I as predictor of both initial and six month outcomes: 0.54+/-0.06 vs. 0.71+/0.07 and 0.53+/- 0.06 vs. 0.65+/-0.07, respectively. The sensitivity of troponin T for both initial and six month
adverse outcomes was significantly lower in the renal group than in the nonrenal group at a similar level of
specificity (0.87): 0.29 vs. 0.60 and 0.45 vs. 0.56, respectively. Troponin I also exhibited similar differences in
sensitivity in the renal group (0.29 vs. 0.50 and 0.33 vs. 0.40, respectively). CONCLUSIONS: The ability of
cardiac troponin T and troponin I to predict risk for subsequent adverse outcomes in patients presenting with
suspected acute coronary syndromes is reduced in the presence of renal insufficiency.

Wilcox G, Archer PD, Bailey M, Dziukas L, Lim CF, Schneider HG. Measurement of cardiac
troponin I levels in the emergency department: predictive value for cardiac and allcause mortality. Medical Journal of Australia 2001;174(4):170-3.
OBJECTIVE: To assess the predictive value of cardiac troponin I levels in cardiac and all-cause mortality in
patients presenting to an emergency department. DESIGN: A prospective cohort study. SETTING: The
emergency department of a major tertiary teaching hospital in metropolitan Melbourne over a six-week period in
1998. PATIENTS: All patients with requests for cardiac enzyme level measurement. MAIN OUTCOME
MEASURES: Cardiac and all-cause mortality within 30 days of presentation. RESULTS: 424 patients (232
men, 192 women; age range, 16-93 years) were reviewed. The 30-day mortality rate was 7.3% (31/424); in
patients with raised levels of both creatine kinase (CK)-MB isoenzyme and troponin I this rate was 27% (7/26;
95% CI, 13%-44%); and in those with troponin I levels above 2 microg/L, but normal CK-MB values, it was
24% (5/21; 95% CI, 5%-43%). The mortality rate in the group with normal results of cardiac markers was 4.3%
(14/328; 95% CI, 2.1%-6.5%). Patients with minor increases in troponin I levels (minimal myocardial damage)
showed an intermediate 30-day mortality rate (13%, 5/39; 95% CI, 2%-24%). Other predictors of 30-day
mortality included age, presentation with shortness of breath, and electrocardiography (ECG) changes
diagnostic of acute myocardial infarction or consistent with ischaemia. Cardiovascular causes were responsible
for most of the deaths in patients with raised troponin I levels. Multivariate logistic regression analysis showed
that raised levels of troponin (> 2.0 microg/L), but not of CK-MB, predict 30-day mortality rate.
CONCLUSIONS: Compared with CK-MB, cardiac troponin I more accurately predicts 30-day mortality rates in
patients presenting to the emergency department. Moreover, troponin I levels identify additional groups of
patients at increased risk of death not so identified by measuring CK-MB values.

Appendix 3: Abstracts of all critically appraised primary publications

69

Myoglobin
Green GB, Skarbek-Borowski GW, Chan DW, Kelen GD. Myoglobin for early risk
stratification of emergency department patients with possible myocardial ischemia.
Academic Emergency Medicine 2000;7(6):625-36.
OBJECTIVES: To determine and compare the prognostic abilities of early, single-sample myoglobin
measurement with that of creatine kinase-MB (CK-MB), with cardiac troponin T (cTnT), and with physician
judgment in the absence of marker results among emergency department (ED) patients with possible myocardial
ischemia. METHODS: Prospective collection of clinical and serologic data using an identity-unlinked technique
from patients with possible myocardial ischemia at two urban EDs. Outcome data concerning the occurrence of
adverse events (AEs) during the 14 days after enrollment were used to calculate and compare the relative risks
(RRs) and predictive values (with 95% confidence intervals) of the various markers for predicting AEs.
RESULTS: Among 396 analyzed patients, 65 (16.4%) accrued 104 AEs, including 13 deaths (3.3%) and 31
(7.8%) myocardial infarctions. Myoglobin predicted AEs (RR = 3.36 [95% CI = 2.19 to 5.15]) with significantly
higher sensitivity (50.8% [95% CI = 38.6 to 62.9]) than either CK-MB (15.4% [95% CI = 6.6 to 24.2]) or cTnT
(24.6% [95% CI = 14.1 to 35.1]), but with lower specificity (81.9% [95% CI = 77.7 to 86.0]; CK-MB = 99.7%
[95% CI = 99.1 to 100]; cTnT = 93.1% [95% CI = 90.3 to 95.8]). Myoglobin had prognostic ability among
patients with chest pain (3.86 [95% CI = 2.39 to 6.22]) and atypical (non-chest pain) presentations (2.71 [95%
CI = 1.09 to 6.71]), including those with a nondiagnostic electrocardiogram (3.11 [95% CI = 1.44 to 6.69]). The
combination of myoglobin and physician decision making identified 63 of the 65 patients (96.9% [95% CI =
92.7 to 100]) with subsequent AEs. CONCLUSIONS: The early prognostic sensitivity of myoglobin may allow
identification of some high-risk patients missed by physician judgment, CK-MB, and cTnT. Myoglobin should
be considered for use in the ED based on both its diagnostic and prognostic abilities.

Other Biomarkers
Bayes-Genis A, Mateo J, Santalo M, et al. D-Dimer is an early diagnostic marker of coronary
ischemia in patients with chest pain. American Heart Journal 2000;140(3):379-84.
BACKGROUND: Chest pain is a frequent symptom in the emergency department and often presents a
diagnostic challenge. Because coronary thrombosis is a hallmark of acute ischemic syndromes, the substrates of
the coagulation and fibrinolysis cascades may be markers of coronary ischemia. The objective of this study was
to determine the diagnostic value of several hemostatic markers in patients presenting to the emergency
department (ED) with chest pain syndromes. METHODS: Two hundred fifty-seven consecutive patients with
acute chest pain were studied in this prospective study conducted in an urban ED. D-Dimer levels were
measured at admission to the ED in all patients. We also measured thrombin-antithrombin complexes,
prothrombin fragment 1+2, activated factor VII, and fibrinogen. We used regression analysis to estimate the
likelihood of myocardial infarction and the diagnostic value of D-dimer. RESULTS: D-Dimer and fibrinogen
levels were significantly higher in patients with acute ischemic events (myocardial infarction and unstable
angina) than in nonischemic patients (P <.01 and P =.02, respectively). The other hemostatic markers were not
significantly elevated in patients with ischemic events. D-Dimer level >500 microg/L had an independent
diagnostic value for myocardial infarction and increased the diagnostic sensitivity of the electrocardiogram and
history from 73% to 92%. CONCLUSION: D-Dimer, an expression of ongoing thrombus formation and lysis, is
a marker of substantial incremental value for the early diagnosis of acute coronary syndromes presenting with
chest pain. It adds independent information to the traditional assessment for myocardial infarction. D-Dimer can
be incorporated into clinical decision models in the ED.

Combination of biomarkers
Huggon AM, Chambers J, Nayeem N, Tutt P, Crook M, Swaminathan S. Biochemical
markers in the management of suspected acute myocardial infarction in the
emergency department. Emerg Med J 2001;18(1):15-9.
OBJECTIVES: To compare cardiac troponin T, myoglobin, CK, CKMB activity, CKMB mass and the initial
electrocardiogram in the early diagnosis of myocardial infarction in the emergency department. MethodsBiochemical markers were measured at presentation in patients with a possible diagnosis of acute myocardial
infarction. Based on the clinical notes, patients were grouped as "definite myocardial infarction" (n = 50),
"definite no myocardial infarction" (n = 81) and "uncertain" (n = 96). Sensitivity and specificity and positive
and negative predictive values were calculated using the 131 patients with definitely present or absent

Appendix 3: Abstracts of all critically appraised primary publications

70

myocardial infarction. RESULTS: The initial electrocardiogram was more sensitive than any of the markers in
the first six hours from symptom onset-sensitivity 74% (95%CI 61% to 88%). The positive predictive value of
the initial electrocardiogram was 97% in the first six hours; the markers ranged from 47% to 67%. The negative
predictive value of the initial electrocardiogram was 85% in the first six hours; the markers ranged from 61% to
70%. Four patients with non-diagnostic electrocardiograms presenting beyond six hours after pain onset had a
myocardial infarct detected by at least three of the biochemical markers in each case. CONCLUSIONS: The
electrocardiogram is of more diagnostic use than biochemical markers in the first six hours after the onset of
pain, but biochemical markers give additional positive diagnostic information in patients presenting later than
this. The negative predictive accuracy of biochemical markers is too low for a single sample to be useful for
excluding myocardial infarction in the first six hours after onset of symptoms.

Jurlander B, Clemmensen P, Wagner GS, Grande P. Very early diagnosis and risk
stratification of patients admitted with suspected acute myocardial infarction by the
combined evaluation of a single serum value of cardiac troponin-T, myoglobin, and
creatine kinase MB(mass). European Heart Journal 2000;21(5):382-9.
AIMS: The diagnostic and prognostic capacity of biochemical markers of acute myocardial infarction in the
emergency department were evaluated in consecutive patients (n=155) with suspected acute myocardial
infarction. METHODS AND RESULTS: Serum myoglobin >/=110 microg. l(-1)and creatine kinase
MB(mass)>/=5 microg. l(-1)had a high accuracy (0.77-0.85) (ns) for acute myocardial infarction diagnosis in
patients presenting >2 h after symptom onset. Troponin-T (>/=0.10 microg. l(-1)) had a lower accuracy (0.530.70) for acute myocardial infarction diagnosis, but was the most important 1-year prognostic marker (cardiac
death or non-fatal acute myocardial infarction). In patients without ST elevation, combined analysis of two
biochemical tests would accurately identify an additional 20% of acute myocardial infarction patients
(predictive value of a positive test=0.82) and also identify those without acute myocardial infarction (predictive
value of a negative test=0.80). One-year event-free survival was excellent (96%) for patients with two negative
biochemical tests, intermediate (74%) for those with discordant tests, and only 53% for patients with two
positive biochemical tests. CONCLUSIONS: Analysis of biochemical tests in the emergency department prior
to hospital admission could accurately identify approximately 20% additional acute myocardial infarction
patients. The prognosis of these patients is poor, and they may be a target for primary PTCA or new early
initiated aggressive medical therapies. Copyright 2000 The European Society of Cardiology.

Ooi SB, Lim YT, Lau TC, Chia BL, Pillai S, Liu T. Value of troponin-T rapid assay, cardiac
enzymes, electrocardiogram and history of chest pain in the initial diagnosis of
myocardial infarction in the emergency department. European Journal of Emergency
Medicine 2000;7(2):91-8.
We conducted a prospective study of 152 adult patients presenting to an emergency department with chest pain
or symptoms suggestive of acute myocardial infarction (AMI) to evaluate the first electrocardiogram (ECG),
creatine kinase (CK)-MB and Troponin-T Rapid Assay (TnT) alone or in combination with chest pain in the
initial diagnosis of AMI. A provisional diagnosis was made after the history, physical examination and the first
ECG reading. Blood specimens were taken for TnT, CK and CK-MB mass. A final discharge diagnosis of AMI
was made according to World Health Organization criteria. Seventy-six (50%) of patients had a final diagnosis
of AMI. The sensitivities of the first ECG, first CK-MB mass and first TnT were 76.3% (95% confidence
interval (CI), 66.8-85.9), 38.2% (95% CI, 27.2-49.1) and 31.6% (95% CI, 21.2-42.0) respectively. The area
under the curve for a combination of ECG, CK-MB mass, TnT and chest pain was the highest at 0.937 when
compared with chest pain with varying combinations of tests. A combination of the first ECG, CK-MB mass
and TnT had a negative predictive value (NPV) of 87.9% (95% CI, 80.0-95.8). The first ECG was the most
sensitive test while the combination of chest pain, ECG, cardiac enzymes and TnT gave the best results in the
initial diagnosis of AMI. If the first ECG, CK-MB mass and TnT are all negative, the probability of having an
AMI is 12%.

Polaczyk CA, Kuntz KM, Sacks DB, Johnson PA, Lee TH. Emergency department triage
strategies for acute chest pain using creatine kinase-MB and troponin I assays: a costeffectiveness analysis. Annals of Internal Medicine 1999;131(12):909-18.
BACKGROUND: Evaluation of acute chest pain is highly variable. OBJECTIVE: To evaluate the costeffectiveness of strategies using cardiac markers and noninvasive tests for myocardial ischemia. DESIGN: Costeffectiveness analysis. DATA SOURCES: Prospective data from 1066 patients with chest pain and from the
published literature. TARGET POPULATION: Patients admitted with acute chest pain. TIME HORIZON:

Appendix 3: Abstracts of all critically appraised primary publications

71

Lifetime. PERSPECTIVE: Societal. INTERVENTIONS: Creatine kinase (CK)-MB mass assay alone; CK-MB
mass assay followed by cardiac troponin I assay if the CK-MB value is normal; CK-MB mass assay followed by
troponin I assay if the CK-MB value is normal and electrocardiography shows ischemic changes; both CK-MB
mass and troponin I assays; and troponin I assay alone. These strategies were evaluated alone or in combination
with early exercise testing. OUTCOME MEASURES: Lifetime cost, life expectancy (in years), and incremental
cost-effectiveness. RESULTS OF BASE-CASE ANALYSIS: For patients 55 to 64 years of age, measurement
of CK-MB mass followed by exercise testing in appropriate patients was the most competitive strategy ($43000
per year of life saved). Measurement of CK-MB mass followed by troponin I measurement had an incremental
cost-effectiveness ratio of $47400 per year of life saved for patients 65 to 74 years of age; it was also the most
cost-effective strategy when early exercise testing could not be performed, CK-MB values were normal, and
ischemic changes were seen on electrocardiography. RESULTS OF SENSITIVITY ANALYSIS: Results were
influenced by age, probability of myocardial infarction, and medical costs. CONCLUSIONS: Measurement of
CK-MB mass plus early exercise testing is a cost-effective initial strategy for younger patients and those with a
low to moderate probability of myocardial infarction. Troponin I measurement can be a cost-effective second
test in higher-risk subsets of patients if the CK-MB level is normal and early exercise testing is not an option.

Porela P, Pulkki K, Helenius H, et al. Prediction of short-term outcome in patients with

suspected myocardial infarction. Annals of Emergency Medicine 2000;35(5):413-20
STUDY OBJECTIVE: Although specific cardiac injury markers have enhanced early patient classification, the
ECG remains a necessary investigation in the acute phase of chest pain. Combined use of both tests could
further improve the diagnostic and prognostic accuracy. METHODS: We studied 311 consecutive patients who
came to the emergency department of a regional referral hospital for the differential diagnosis of acute chest
discomfort. The admission ECG was classified using an automated interpretation program and tested together
with elevated admission creatine kinase isoform MB (CK-MB) and cardiac troponin I (TnI) concentration for
prediction of final myocardial injury (44%) and in-hospital mortality (14%). RESULTS: Combining the
information from the admission ECG and cardiac markers, the sensitivity for becoming final myocardial injury
(maximal CK-MB >/=11 microg/L) was 90% and specificity 61%. The proportion of false-negative results
(10%) was independent of symptom duration. Age, positive ECG findings, and increased admission TnI levels
were predictive for in-hospital mortality. CONCLUSION: The commonly available biochemical and ECG
criteria allow risk stratification of patients with a suspected acute ischemic event. The data analysis can easily be
automated and is independent of patient delay.

Wu AH, Ghani F, Prigent F, Petry C, Armstrong G, Graff L. Reflex testing II: evaluation of
an algorithm for use of cardiac markers in the assessment of emergency department
patients with chest pain. Clinica Chimica Acta 1999;288(1-2):97-109.
A reflex algorithm was developed and evaluated for the use of serum cardiac markers for the diagnosis and rule
out of acute myocardial infarction (AMI), and risk stratification of unstable angina patients for those who
present to emergency departments (ED) with chest pain. The process begins with testing of total CK and
myoglobin at admission. Based on these results, the algorithm determines the need for subsequent testing for the
CK-MB isoenzyme and cardiac troponin I (cTnI). The algorithm also directs the need for further blood
collection and cardiac marker testing at 4, 8, and 12 h after presentation. A total of eleven stopping points were
identified. For some of these stopping points, the algorithm concluded that further blood collections and testing
was unnecessary and redundant. The algorithm was retrospectively evaluated on 101 non-consecutive chest pain
patients who presented to the EDs at three hospitals. For the AMI group (n=34), six of nine possible different
stopping points were reached: 64.7% of cases were diagnosed with the first sample at admission, an additional
32.3% after 4 h, and 2.9% at 8 h. The 12-h sample was not necessary for any of the AMI patients. For the nonAMI group (n=67), most reached the stopping point of no cardiac injury or risk. There were five unstable angina
patients who had minor myocardial damage on the basis of a marginally increased cTnI. Of these, one patient
subsequently suffered AMI, and three others required angioplasty or bypass surgery. Compared to performing
four tests on all patient samples, the reflex algorithm would have reduced the number of necessary tests from
442 to 130 (71% reduction) for AMI patients, and 871 to 469 (46% reduction) for non-AMI patients, if
prospectively implemented.

Zimmerman J, Fromm R, Meyer D, et al. Diagnostic marker cooperative study for the
diagnosis of myocardial infarction. Circulation 1999;99(13):1671-7.
BACKGROUND: Millions of patients present annually with chest pain, but only 10% to 15% have myocardial
infarction. Lack of diagnostic sensitivity and specificity of clinical and conventional markers prevents or delays
treatment and leads to unnecessary costly admissions. Comparative data are lacking on the new markers, yet

Appendix 3: Abstracts of all critically appraised primary publications

72

using all of them is inappropriate and expensive. METHODS AND RESULTS: The Diagnostic Marker
Cooperative Study was a prospective, multicenter, double-blind study with consecutive enrollment of patients
with chest pain presenting to the emergency department. Diagnostic sensitivity and specificity and frequency of
increase in patients with unstable angina were determined for creatine kinase-MB (CK-MB) subforms,
myoglobin, total CK-MB (activity and mass), and troponin T and I on the basis of frequent serial sampling for
</=24 hours. Of 955 patients with chest pain, 119 (12.5%) had infarction identified by use of CK-MB mass, and
203 (21%) had unstable angina. CK-MB subforms were most sensitive and specific (91% and 89%) within 6
hours of onset, followed by myoglobin (78% and 89%). For late diagnosis, total CK-MB activity (derived from
subforms) was the most sensitive and specific (96% and 98%) at 10 hours from onset, followed by troponin I
(96% and 93%), but not until 18 hours, and troponin T (87% and 93% at 10 hours). In unstable angina, CK-MB
subforms were increased in 29.5%, myoglobin in 23.7%, troponin I in 19.7%, and troponin T in 14.8%. All
markers were increased in 99 patients. With each marker as the diagnostic standard, CK-MB subforms and
myoglobin remained the most sensitive for early diagnosis. CONCLUSIONS: The CK-MB subform assay alone
or in combination with a troponin reliably triages patients with chest pain and should lead to improved therapy
and reduced cost

Grzybowski M, Zalenski RJ, Ross MA, Bock B. A prediction model for prehospital triage of
patients with suspected cardiac ischemia. Journal of Electrocardiology
2000;33(Suppl):253-8.
The American College of Cardiology recommends that patients with high risk acute myocardial infarction
(AMI) be triaged to hospitals with percutaneous transluminal coronary angioplasty capability. However, there
are no prehospital triage criteria to select candidates for bypassing community hospitals and being taken directly
to "cardiac centers." This article assesses which independent variables predict death within 7 days in patients
with suspected AMI transported by EMS. This is a retrospective study of 291 AMI patients transported by
ambulance to 3 hospitals during 1996-1997. Included were patients who were (n = 244) > or =18 years of age,
had a ED chief complaint of chest pain or dyspnea for whom we had mortality data. Mortality at 7 days, our
primary outcome measure, was obtained by using a metropolitan Detroit tricounty death index records.
Differences between the survivors and nonsurvivors were assessed using the Student's t-test and chi-square tests.
Multiple triage criteria were assessed for optimal identification of high risk patients by constructing a logistic
multivariate model. Among the study population, 15% died within 7 days (95% confidence interval (CI) 10.319.2), and this group represented 63.2% of all deaths over a 2 year surveillance period. Survivors, compared to
nonsurvivors, were 14.1 years younger (P < or = .001) and more often men (P < or = 0.001). The dispatch time
to ED arrival was less among survivors than nonsurvivors (42.8 vs. 50.6 min, P < or = .01). EMS vital signs
differed by survivor status. Among survivors, HR was lower (-11.9 bpm; P < or = 0.01), RR was lower (-6.7
rpm; P < or = .001), SBP was higher (+14.5 mmHg; P < or = 0.05) and DBP was higher (+13.2 mm Hg; P < or
= .01). A multivariate model identified the following as independent predictors of early mortality: female gender
(OR = 2.3; P < or = .05), age > or =65 (OR = 5.9; P < or = .01), RR > or = 20 (OR = 4.6; P < or = .001), SBP <
120 (OR = 2.4; P < or = .05). The overall model was 86% sensitive and 53% specific with an area under the
receiving operating characteristic curve of 0.8 (P < or = .001). A triage rule based on a multivariate model can
identify the group at high risk of early cardiac death. This decision rule needs to be prospectively validated.

Echocardiography
No new data

Technetium-99m Sestamibi Myocardial Perfusion Imaging

Kontos MC, Jesse RL, Anderson FP, Schmidt KL, Ornato JP, Tatum JL. Comparison of
myocardial perfusion imaging and cardiac troponin I in patients admitted to the
emergency department with chest pain. Circulation 1999;99(16):2073-8.
BACKGROUND: Identification of patients with acute coronary syndromes (ACS) among those who present to
emergency departments with possible myocardial ischemia is difficult. Myocardial perfusion imaging with
99mTc sestamibi and measurement of serum cardiac troponin I (cTnI) both can identify patients with ACS.
METHODS AND RESULTS: Patients considered at low to moderate risk for ACS underwent gated singlephoton emission CT sestamibi imaging and serial myocardial marker measurements of creatine kinase-MB, total
creatine kinase activity, and cTnI over 8 hours. Positive perfusion imaging was defined as a perfusion defect

Appendix 3: Abstracts of all critically appraised primary publications

73

with associated abnormalities in wall motion or thickening. cTnI >/=2.0 ng/mL was considered abnormal.
Among the 620 patients studied, 59 (9%) had myocardial infarction and 81 (13%) had significant coronary
disease; of these patients, 58 underwent revascularization. Perfusion imaging was positive in 241 patients
(39%), initial cTnI was positive in 37 (6%), and cTnI was >/=2.0 ng/mL in 74 (12%). Sensitivity for detecting
myocardial infarction was not significantly different between perfusion imaging (92%) and cTnI (90%), and
both were significantly higher than the initial cTnI (39%). Sensitivity for predicting revascularization or
significant coronary disease was significantly higher for perfusion imaging than for serial cTnI, although
specificity for all end points was significantly lower. Lowering the cutoff value of cTnI to 1.0 ng/mL did not
significantly change the results. CONCLUSIONS: Early perfusion imaging and serial cTnI have comparable
sensitivities for identifying myocardial infarction. Perfusion imaging identified more patients who underwent
revascularization or who had significant coronary disease, but it had lower specificity. The 2 tests can provide
complementary information for identifying patients at risk for ACS.

Kosnik JW, Zalenski RJ, Grzybowski M, Huang R, Sweeny PJ, Welch RD. Impact of
technetium-99m sestamibi imaging on the emergency department management and
costs in the evaluation of low-risk chest pain. Academic Emergency Medicine
2001;8(4):315-23.
OBJECTIVES: To assess the impact of rest sestamibi scanning on emergency physicians' (EPs') diagnostic
certainty and decision making (as assessed by the hypothetical disposition of patients) for 69 consenting stable
patients with suspected acute cardiac ischemia and nondiagnostic electrocardiograms. The resultant impact on
costs was examined as a secondary outcome. METHODS: Patients with suspected acute cardiac ischemia were
injected with 25 mCi of sestamibi within two hours of active pain in one of three emergency department study
sites. The probability of acute myocardial infarction (AMI) and unstable angina (UA), and hypothetical
disposition decisions were recorded immediately before and after physicians were notified of scan results.
Changes in disposition were classified as optimal or suboptimal. For the cost determinations, a cost-based
decision support program was used. RESULTS: For the subgroup found to be free of acute cardiac events
(ACEs) (n = 62), the EPs' post-sestamibi scan probabilities for AMI decreased by 11% and UA by 18% (p <
0.001 for both conditions). In seven patients with ACEs, the post-scan probabilities of AMI and UA increased,
but neither was statistically significant. Scan results led to hypothetical disposition changes in 29 patients (42%),
of which 27 (93%) were optimal (nine patients were reassigned to a lower level of care, two to a higher level,
and 16 additional patients to "discharge-home" status). The strategy of scanning all patients who were low to
moderate risk for acute cardiac ischemia would result in an increase of direct costs of care of $222 per patient
evaluated, due to added cost of sestamibi scanning. CONCLUSIONS: Sestamibi scanning results appropriately
affected the EPs' estimates of the probability of AMI and UA and improved disposition decisions. Scanning all
low-risk patients would likely be associated with increased costs at this medical center.

Mariani G, Villa G, Rossettin PF, et al. Detection of acute myocardial infarction by 99mTclabeled D-glucaric acid imaging in patients with acute chest pain. Journal of Nuclear
Medicine 1999;40(11):1832-9.
Definitive diagnosis of acute myocardial infarction early in the process is often difficult. An imaging agent that
localized quickly and specifically in areas of acute necrosis could provide this critical diagnostic information.
To determine whether imaging with 99mTc-labeled D-glucaric acid (GLA) could provide this information, we
imaged a group of patients presenting with symptoms suggestive of acute infarction. METHODS: Twenty-eight
patients presenting to the emergency department with symptoms highly suggestive of acute infarction were
injected with 99mTC-GLA and imaged about 3 h later. RESULTS: The sensitivity of lesion detection was
remarkably time dependent. Fourteen patients with acute infarction injected within 9 h of onset of chest pain had
positive scans, even in the presence of persistent occlusion. The remaining 14 patients had negative scans. Nine
patients with negative scans had acute infarction but were injected more than 9 h after onset of chest pain. The
final diagnosis in the remaining 5 patients was unstable angina (3 injected <9 h and 2 injected >9 h after onset of
chest pain). Six patients were reinjected with 99mTc-GLA 4-6 wk after their initial study to determine whether
persistent positive scans occurred with this agent. All 6 had negative scans. CONCLUSION: This study
suggests that 99mTc-GLA localizes in zones of acute myocardial necrosis when injected within 9 h of onset of
infarction.

Appendix 3: Abstracts of all critically appraised primary publications

74

Paventi S, Parafati MA, Di Luzio E, Pellegrino CA. Safety and feasibility of two-dimensional
echocardiography and myocardial perfusion imaging in patients with chest pain.
Angiology 2001;52(5):305-9.
The accurate identification of patients at high risk for acute coronary syndromes among those seen in the
emergency department with possible myocardial ischemia and nonischernic electrocardiograms is a problem.
Two-dimensional echocardiography and myocardial perfusion imaging with 99m-technetium sestamibi can
identify patients at low and high risk; however, comparative studies are lacking. The authors studied 555
patients considered at low or moderate risk for myocardial ischemia in our emergency department on the basis
of the presenting history, and results of physical examination and electrocardiography. These patients underwent
echocardiography and myocardial perfusion imaging within 4 hours of presentation. Endpoints included
myocardial infarction, percutaneous transluminal coronary angioplasty, and positive results on stress perfusion
imaging. Both imaging procedures were performed in the emergency department on 370 patients. Overall
agreement between the two techniques was high (concordance, 89%) in the patients who had myocardial
infarction or underwent coronary angiography. Agreement between the two techniques is high when used in
patients with possible myocardial ischemia. Both techniques helped identify patients at high risk who required
admission and those who could be safely discharged.

Electron beam computed tomography

Laudon DA, Vukov LF, Breen JF, Rumberger JA, Wollan PC, Sheedy PF, 2nd. Use of
electron-beam computed tomography in the evaluation of chest pain patients in the
emergency department. Annals of Emergency Medicine 1999;33(1):15-21.
Study objective: We sought to determine whether electron-beam computed tomography (EBCT) could be used
as a triage tool in the emergency department for patients with angina-like chest pain, no known history of
coronary disease, normal or indeterminate ECG findings, and normal initial cardiac enzyme concentrations.
METHODS: We conducted a prospective observational study of 105 patients admitted between December 1995
and October 1997 to the ED of a large tertiary care hospital with 70,000 annual ED visits. The study group was
comprised of women aged 40 to 65 years and men aged 30 to 55 years who presented with angina-like chest
pain requiring admission to the hospital or chest pain observation unit. All patients underwent EBCT of the
coronary arteries, along with other cardiac testing as deemed necessary by staff physicians. RESULTS: Of the
105 patients, 100 underwent other cardiac testing during hospitalization. Evaluation included treadmill exercise
testing in 58, coronary angiography in 25, radionuclide stress testing in 19, and echocardiography in 11. Results
of EBCT and cardiac testing were negative for both in 53 patients (53%), positive for both in 14 (14%), positive
for tomography and negative for cardiac testing in 32 (32%), and negative for tomography and positive for
cardiac testing in only 1 patient. This positive test result, on a treadmill exercise test, was ruled a false positive
by an independent staff cardiologist. Two other female patients with normal exercise sestamibi or coronary
angiography and EBCT findings also had false-positive treadmill exercise results. The sensitivity of EBCT was
100% (95% confidence interval, 77% to 100%), with a negative predictive value of 100% (95% confidence
interval, 94% to 100%). Specificity was 63% (95% confidence interval, 54% to 75%). CONCLUSION: EBCT is
a rapid and efficient screening tool for patients admitted to the ED with angina-like chest pain, normal cardiac
enzyme concentrations, indeterminate ECG findings, and no history of coronary artery disease. Our study
suggests that patients with normal initial cardiac enzyme concentrations, normal or indeterminate ECG findings,
and negative results on EBCT may be safely discharged from the ED without further testing or observation.
Larger studies are required to confirm this conclusion.

Acute Cardiac Ischemia Time-Insensitive Predictive Instrument

(ACI-TIPI)
No new data.

Appendix 3: Abstracts of all critically appraised primary publications

75

Goldman Chest Pain Protocol

Durairaj L, Reilly B, Das K, et al. Emergency department admissions to inpatient cardiac
telemetry beds: a prospective cohort study of risk stratification and outcomes.
American Journal of Medicine 2001;110(1):7-11.
PURPOSE: Little is known about physicians' use of inpatient cardiac telemetry units among emergency
department patients at risk for cardiac complications. We therefore studied the outcomes of patients admitted to
inpatient telemetry beds to identify a subset of patients from whom cardiac monitoring could be withheld safely.
SUBJECTS AND METHODS: We conducted a prospective cohort study of 1, 033 consecutive adult patients
admitted to an inpatient telemetry unit from the emergency department of a 700-bed urban public teaching
hospital. Subjects with or without chest pain were risk-stratified using a prediction rule and observed for inhospital cardiac complications, acute myocardial infarction, and transfer to an intensive care unit (ICU).
RESULTS: There were no significant differences between patients with (n = 677) or patients without chest pain
(n = 356) in the rates of major cardiac complications, myocardial infarctions, or transfers to an ICU. Among 318
patients with chest pain who were classified as being very low risk, none suffered major complications (negative
predictive value 100%; 95% confidence interval [CI]: 98.8% to 100%). Among 214 very low risk patients
without chest pain, 1 (0.5%) had a major complication (negative predictive value 99.5%; 95% CI: 97.4% to
99.9%). CONCLUSIONS: The prediction rule accurately identified patients with or without chest pain who
were at very low risk of major complications, identifying a subset from whom cardiac monitoring could be
withheld safely.

Reilly B, Durairaj L, Husain S, et al. Performance and potential impact of a chest pain
prediction rule in a large public hospital. American Journal of Medicine
1999;106(3):285-91.
PURPOSE: To evaluate the performance of a previously validated prediction rule for patients presenting to the
emergency department with chest pain and the potential impact of the rule on triage decisions. SUBJECTS AND
METHODS: In a prospective cohort study, physician investigators interviewed consecutive patients admitted for
suspected acute ischemic heart disease (n = 207) by emergency department attending physicians who had not
used the prediction rule. We measured the accuracy of the rule in predicting cardiac complications in these
patients, and compared actual triage decisions with those that might have been recommended by use of the
prediction rule. We also measured comorbid illnesses among patients stratified as very low risk by the
prediction rule, as well as the effect of standardizing the definition of unstable angina and interpretation of
electrocardiograms (ECG) on the rule's sensitivity and specificity. RESULTS: Overall, the rate of major cardiac
complications (4.3%) was similar to that reported in the original study (3.6%). The prediction rule performed
well in predicting these complications in our patients (area under receiver operating characteristic curve 0.84
versus 0.80 in the original study; difference 0.04, 95% confidence interval [CI] -0.07, 0.14). Standardized
definitions of unstable angina and interpretation of ECGs improved the specificity of the prediction rule in
predicting complications (55% versus 47%; difference 8%, 95% CI 1.5%, 13.7%). The prediction rule
recommended admission to telemetry units in 65 fewer patients than actually occurred (31% of the entire
cohort). None of these patients had major complications. A substantial minority of "very low risk" patients
(27%) had comorbid illnesses requiring inpatient treatment. CONCLUSIONS: This independent validation of
the prediction rule suggests that it can improve triage decisions for patients admitted with suspected acute
ischemic heart disease. Additional studies are needed to test prospectively the performance of the prediction rule
in actual decision making, its acceptance by clinicians, and its cost effectiveness.

Other Computer-Based Decision Aids

Aase O. Clinical experience with a decision support computer program using Bayes' theorem
to diagnose chest pain patients. Cardiology 1999;92(2):128-34.
A decision support computer program (DSP) was used by the emergency room physician as a diagnostic tool
on patients admitted with acute chest pain to guide the referral of these patients either to the Coronary Care Unit
(CCU) or general ward. The DSP used Bayes' theorem on 38 anamnestic and clinical variables to classify
patients into one of nine diagnoses. During a six months trial period 32 physicians used the DSP to diagnose 493
patients admitted with acute chest pain. The physicians referred the patients to CCU or general ward based on
their clinical judgements, the ECG findings and the diagnostic estimates given by the DSP. The program
correctly diagnosed 150 (84%) of 178 patients with acute myocardial infarction and 63 of 112 patients with
unstable angina. However, acute ischemic heart disease (acute myocardial infarction or unstable angina) was

Appendix 3: Abstracts of all critically appraised primary publications

76

correctly classified by the DSP for 259 (89%) of 290 patients. By using the DSP, the number of patients
unnecessarily referred to CCU was reduced from 35% to 19% and the number of patients in need of CCU
observation misallocated to general ward was reduced from 13% to 10%. Thus, use of the DSP in the
emergency room on easily available anamnestic and clinical variables may improve referrals to the CCU,
optimize therapy and resource use. Copyright 2000 S. Karger AG, Basel

Left bundle-branch block subpopulation specific studies

Kontos MC, McQueen RH, Jesse RL, Tatum JL, Ornato JP. Can myocardial infarction be
rapidly identified in emergency department patients who have left bundle-branch
block? [see comments]. Annals of Emergency Medicine 2001;37(5):431-8.
STUDY OBJECTIVES: Fibrinolytic therapy is recommended for patients who have chest pain and left bundlebranch block (LBBB). However, the presence of baseline ECG abnormalities makes early accurate identification
of acute myocardial infarction (AMI) difficult. The predictive ability of clinical and ECG variables for
identifying patients with LBBB and AMI has not been well studied. We sought to determine the prevalence and
predictors of myocardial infarction among patients presenting to the emergency department with LBBB on the
initial ECG who were evaluated for myocardial infarction. METHODS: All patients presenting to the ED were
prospectively risk stratified on the basis of clinical and historical variables. ECGs from patients with LBBB
were compared retrospectively with previously published criteria for identification of AMI. The ability of a new
LBBB to predict AMI was also determined. RESULTS: Twenty-four (13%) of the 182 patients with LBBB had
AMI. Clinical and historical variables were similar in patients with and without AMI. A new LBBB had a
sensitivity of 42% and a specificity of 65%. The presence of concordant ST-segment elevation or depression had
specificities and positive predictive values of 100%; however, sensitivities were only 8% and 17%, respectively.
The best diagnostic criterion was the presence of concordant ST-segment elevation or depression on the ECG or
an initially elevated creatine kinase MB (sensitivity, 63%; specificity, 99%). CONCLUSION: ECG criteria for
identifying patients with AMI and LBBB identify only a small minority of patients with AMI. Treating all
patients with LBBB and chest pain with fibrinolytics would result in treatment of a significant number of
patients without AMI.

Shlipak MG, Lyons WL, Go AS, Chou TM, Evans GT, Browner WS. Should the
electrocardiogram be used to guide therapy for patients with left bundle-branch block
and suspected myocardial infarction? [see comments]. JAMA 1999;281(8):714-9.
CONTEXT: Recently, an algorithm based on the electrocardiogram (ECG) was reported to predict myocardial
infarction (MI) in patients with left bundle-branch block (LBBB), but the clinical impact of this testing strategy
is unknown. OBJECTIVE: To determine the diagnostic test characteristics and clinical utility of this ECG
algorithm for patients with suspected MI. DESIGN: Retrospective cohort study to which an algorithm was
applied, followed by decision analysis regarding thrombolysis made with or without the algorithm. SETTING:
University emergency department, 1994 through 1997. PATIENTS: Eighty-three patients with LBBB who
presented 103 times with symptoms suggestive of MI. MAIN OUTCOME MEASURES: Myocardial infarction
determined by serial cardiac enzyme analyses and stroke-free survival. RESULTS: Of 9 ECG findings assessed,
none effectively distinguished the 30% of patients with MI from those with other diagnoses. The ECG algorithm
indicated positive findings in only 3% of presentations and had a sensitivity of 10% (95% confidence interval,
2%-26%). The decision analysis showed that among 1000 patients with LBBB and chest pain, 929 would
survive without major stroke if all received thrombolysis compared with 918 if the ECG algorithm was used as a
screening test. CONCLUSIONS: The ECG is a poor predictor of MI in a community-based cohort of patients
with LBBB and acute cardiopulmonary symptoms. Acute thrombolytic therapy should be considered for all
patients with LBBB who have symptoms consistent with MI.

Appendix 3: Abstracts of all critically appraised primary publications

77

Appendix 4 Disclaimer and NHMRC Levels of Evidence

Disclaimer
The information in this report is a summary of the available evidence in excluding acute
cardiac ischaemic events in patients presenting to emergency departments with chest pain.
The report is designed to give the reader a starting point for considering the research evidence
and for continuing to search out further evidence. Whilst great care has been taken in the
preparation of the materials included in this publication, the authors and Melbourne Health do
not warrant the accuracy of this document, any representation, implied or expressed,
concerning the efficacy, appropriateness or suitability of any treatment or product. In view of
the possibility of human error or further advances in medical knowledge the authors and
Melbourne Health cannot and do not warrant that the information contained in these pages is
in every aspect accurate and complete. Accordingly, they are not and will not be held
responsible or liable for any errors of omissions that may be found in this report. You are
therefore encouraged to consult other sources in order to confirm the information contained
in this report and, in the event that medical treatment is required, to take professional expert
advice from a legally qualified and appropriately experienced medical practitioner.
Levels of evidence used in searching for research material by the Clinical Epidemiology
and Health Service Evaluation Unit
As defined by How to use the evidence: assessment and application of scientific evidence
from the National Health and Medical Research Council, Canberra 2000.
Level

Type of Evidence

Evidence obtained from a systematic review (or meta-analysis)

of all relevant randomised controlled trials.

II

Evidence obtained from at least one randomised controlled trial.

III - 1

Evidence obtained from pseudorandomised controlled trials

(alternate allocation or some other method).

III - 2

Evidence obtained from comparative studies (including

systematic reviews of such studies) with concurrent controls
and allocation not randomized, cohort or case-control studies or
interrupted time series with a control group.

III - 3

Evidence obtained from comparative studies with historical

control, two or more single-arm studies or interrupted time
series without a parallel control group.

IV

Evidence obtained from case series, either post-test or

pretest/post test.

78