Clinical Features, Triage, and Outcome of

Clinical Features, Triage, and Outcome of
Patients Presenting to the ED With Suspected
Acute Coronary Syndromes but Without Pain:
A Multicenter Study
BORIS E. CORONADO, MD,* J. HECTOR POPE, MD,*† JOHN L. GRIFFITH, PHD,*
JONI R. BESHANSKY, RN, MPH,* AND HARRY P. SELKER, MD, MSPH*
We studied the impact on triage and outcome of patients presenting to
the emergency department (ED) with symptoms suggestive of an acute
coronary syndromes (ACS) but without a complaint of pain. Data from a
prospective clinical trial of patients with symptoms suggesting an ACS
in the EDs of 10 US hospitals comparing patient demographics, clinical
variables, and outcomes was used to perform a secondary analysis. Of
10,783 subjects, a final diagnosis of an ACS was confirmed in 24% of
which 35% had acute myocardial infarction (AMI) and 65% unstable
angina pectoris (UAP). Pain was absent in 6.2% of patients with acute
ischemia and in 9.8% of those with AMI. Compared to similar patients
who presented with pain, patients with painless ischemia were older,
were more commonly women, had more cardiac and related diseases.
Among patients with AMI, fewer patients without pain were admitted to
critical care units compared to similar patients with pain. Among patients
with AMI, logistic regression predicting lack of pain identified age, heart
failure and diabetes, with only age and heart failure among all with ACS.
After controlling for clinical features, lack of pain during acute ischemia
predicted increased hospital mortality. We concluded that age and heart
failure are independently associated with painless ACS, in addition to
diabetes among those with AMI. Lack of pain predicts increased hospital
mortality in patients with ACI through mechanisms that remain to be
elucidated. There is a need for greater awareness in the general public of
the different manifestations of ACS to enhance the recognition and
prompt response to their symptoms. mortality (Am J Emerg Med 2004;
22:568-574. © 2004 Elsevier Inc. All rights reserved.)
Since the classic description of angina pectoris by Heberden in 1768,1 chest pain has been recognized as the most
common clinical manifestation of an acute coronary syn-
From the *Center for Cardiovascular Health Services Research,
Institute for Clinical Research and Health Policy Studies, Department of Medicine, Tufts-New England Medical Center, Tufts University School of Medicine, Boston, Massachusetts and †Department of Emergency Medicine, Baystate Medical Center, Springfield,
MA.
Manuscript received July 26, 2003; accepted November 24, 2003.
Presented at the American Federation of Medical Research Annual Meeting, Spring, 1995 abstract (J Invest Med 1995;43:503A)
Supported by the Agency for Health Care Policy and Research
grant nos. T32HS00060 and nos. RO1HS07360 and the National
Heart, Lung and Blood Institute grant no. R01 HL53900.
Address reprint requests to Harry P. Selker, MD, MSPH, Institute
for Clinical Research and Health Policy Studies, Tufts, New England
Medical Center, #63, 750 Washington Street, Boston, MA 02111.
Email: [email protected]
Key Words: Acute coronary syndrome, acute cardiac ischemia,
acute myocardial infarction, unstable angina pectoris, chest pain,
clinical presentation, emergency department, diagnosis, triage,
© 2004 Elsevier Inc. All rights reserved.
0735-6757/04/2207-0013$30.00/0
doi:10.1016/j.ajem.2004.09.001
568
dromes (ACS): unstable angina pectoris (UAP) and acute
myocardial infarction (AMI). Although the World Health
Organization2 requires the presence of chest pain as one of
the cornerstone features of its definition of AMI, a new
definition AMI from the European Society of Cardiology
and the American College of Cardiology (ESC/ACC) requires only a history of ischemic symptoms and a typical
rise and fall of at least one of the biochemical markers.3
Not all patients with ACS present with chest pain and the
extent to which this phenomenon occurs is largely unknown. Some investigators have found that up to 30% of
patients with ACS may not experience this symptom,4-11
and many experience no pain whatsoever12 These patients
often complain of shortness of breath, extreme fatigue,
nausea or fainting.9 Barron et al13 reported that the absence
of chest pain at presentation was among the most significant
factors predicting lower use of thrombolytic therapy among
a subset of AMI patients eligible for such treatments in the
National Registry of Myocardial Infarction 2 (NRMI-2).
Whether these patients are also less likely to receive other
important treatments in the management of AMI remains
unclear.
These findings have heightened the interest in the study
of the mechanisms of pain perception during myocardial
ischemia and in the processes that lead to defective pain
perception in certain individuals.14 However, this population has not been well characterized. While some epidemiologic studies identified women and advanced age in men as
factors associated with unrecognized AMI,5 other reports
identified diabetes mellitus11,15 and hypertension.16,17 These
data remain controversial and other clinical factors remain
undefined due in part to a paucity of studies with multivariable analyses with proper adjustment for confounders.18 A
better understanding of the factors associated with atypical
presentations of an ACS will eventually help optimize the
early recognition and treatment of these patients when they
become symptomatic.
Finally, little is known about how the absence of pain
affects clinical outcomes in patients who present to the
emergency department (ED) with otherwise symptomatic
ischemia. There is ongoing controversy not only as to which
patients experience painless ischemia more commonly but
also how this is related to short- and long-term mortality
rates.19-30 While lack of chest pain during AMI has been
linked to higher mortality rates,12,31 there is a need for
studies of prospectively collected data on all patients with
ACS and not only those with confirmed infarctions. This is
CORONADO ET AL ■ ACS WITHOUT PAIN
particularly important since ACS is a dynamic continuum
with ST elevation myocardial infarction (STEMI) at one
end of the ischemic cascade and non- ST elevation ACS at
the other, and since prompt diagnosis and treatment may
themselves determine whether or not myocardial necrosis
occurs.
To gain insight into the clinical features of patients with
symptoms suggesting an ACS but without pain as well as to
study the impact of this clinical presentation on ED triage
(admit/discharge decision) and outcomes (30-day mortality), we studied patients included in a large prospective
study of ED patients undergoing evaluation for a suspected
ACS.29
MATERIALS AND METHODS
Study Subjects
The study included all consenting adults ages 30 and
older who presented to the ED at any of the participating
hospitals with symptoms suggestive of an ACS who were
entered in the prospective multicenter Acute Cardiac Ischemia-Time Insensitive Predictive Instrument (ACI-TIPI)
Clinical Trial.29 These symptoms, described in the Imminent Myocardial Infarction Rotterdam Study,30 included
chest pain, chest pressure, left arm pain, jaw pain, upper
abdominal pain, dizziness, nausea, vomiting, or dyspnea.
Additional symptoms were recorded on all patients with
suspected cardiac ischemia. Patients in whom a non-cardiac
cause for their symptoms could not be established in the ED
were included in the study.
In the original trial every patient’s presenting symptoms
were classified as primary (chief complaint) or secondary.
For the purposes of this study, we defined a clinical presentation as painful if pain was reported by study subjects
regardless of anatomic location, severity, duration, associated symptoms, or whether pain was the primary or secondary complaint. Conversely, a painless presentation was one
in which pain was not elicited from the patient’s history
prior to or upon arrival to the ED. These patients often
complained of shortness of breath, extreme fatigue, nausea
or fainting.9
Patients with chest pain of clear traumatic origin or other
obvious noncardiac cause were excluded, as were those
transferred from other institutions and those in whom consent was declined or unobtainable by virtue of mental illness
or language barrier. Patients below the age of 30 in whom
cocaine abuse was suspected, as well as those who presented to the ED with cardiac arrest, though included in the
original trial, were excluded from this study. The overall
patient inclusion rate in the original trial was 92%.
Final diagnoses were assigned by study site physician
blinded review (with 10% overreading by Study Center to
confirm interrater reliability) using World Health Organization criteria2 based on presentation and clinical course,
initial and follow-up ECGs, and creatine kinase and
CK-MB levels. All enrolled patients had at least one ECG
and serial creatine kinase and CK-MB determinations. Cardiac diagnoses were considered ischemic or non-ischemic.
Cardiac ischemia was divided into AMI and other acute
coronary syndromes, with the former further classified according to the Killip classification.31 Angina severity was
569
rated both by Canadian Cardiovascular Society (CCS) classification criteria32 and by time duration since symptom
onset or worsening of anginal pattern. Unstable angina
pectoris (UAP) was defined as CCS Class 3 with new or
changed symptoms for less than 3 days or CCS Class 4
symptoms of any duration; stable angina was defined as
CCS Class 1 or 2, or Class 3 angina with no change in
symptoms within 3 days. A final diagnosis of angina pectoris or acute ischemia without infarction was made based
on clinical presentation, electrocardiographic changes and
other confirmatory tests performed after initial ED evaluation, such as exercise electrocardiography tests with or
without nuclear scintigraphy or coronary angiography.
Other cardiac and noncardiac diagnoses were assigned in a
similar fashion after appropriate diagnostic investigation
and thorough review of the patient’s medical record. ED
triage was analyzed according to whether patients were
discharged to home or admitted to a critical care unit,
monitored bed or nonmonitored bed.
Study Sites
The 10 participating hospitals included Baystate Medical
Center (Springfield, MA), Boston City Hospital, Boston
University Medical Center, Medical College of Virginia
(Richmond, VA), Medical College of Wisconsin (Milwaukee, WI), New England Medical Center (Boston, MA),
Newton-Wellesley Hospital (Newton, MA), Rhode Island
Hospital (Providence, RI), University of Cincinnati Medical
Center, and University of North Carolina Hospitals (Chapel
Hill, NC). These hospitals represented a range of settings
and practice types, serving patients from different sociodemographic and ethnic backgrounds, from impoverished urban areas to affluent suburbs. Hospital sizes ranged from
350 to 1,019 beds, with yearly ED visits varying from
27,000 to 115,000. Institutional review boards of all participating hospitals approved the study.
Data Collection
During 7 months starting in May 1993, data were collected prospectively and concurrently upon ED presentation, during hospitalization, and at 30-day follow-up. These
included sociodemographic information, initial and follow-up clinical features, first 12-lead ECG, and serial creatine kinase and CK-MB tests. Patients discharged from the
ED returned within 24-72 hours for repeat clinical evaluation, 12-lead ECG, and creatine kinase and CK-MB testing.
The follow-up rate for data needed for definitive assignment
of a diagnosis, including those not hospitalized, was 99%.
Data Analysis
The analysis of painless ischemia was approached first by
identifying presenting symptoms in patients without pain
and subsequently by determining the impact of such manifestations on hospital mortality.
Statistical Methods
The aim of this study is to compare patient demographics
(age, gender), clinical variables (absence for pain, initial ED
systolic blood pressure, heart failure, past medical history of
570
AMERICAN JOURNAL OF EMERGENCY MEDICINE ■ Volume 22, Number 7 ■ November 2004
diabetes mellitus), and outcomes (final diagnosis, admissions rates, and 30-day mortality) between patients with a
suspected ACS both with and without ischemic pain as a
complaint.
All hypothesis tests were 2-tailed. Analyses of differences in baseline demographic and clinical characteristics
between groups used the Student’s t-test for continuous
variables and Chi-square tests for dichotomous variables.
Hospital mortality was defined as deaths occurring in the
hospital at any moment from arrival to the ED through
hospital discharge. Deaths occurring in the ED were
counted as hospital deaths.
To determine factors that may be associated with painless
ACS, we constructed a multivariable logistic regression
model with painless ischemia as the outcome variable. Candidate variables were allowed to enter the model in a stepwise fashion. Clinically relevant factors such as age, radiographic evidence of congestive heart failure on presentation,
diabetes, gender, history of hypertension and prior infarction were initially tested by univariable analysis. Of these,
those with unadjusted statistical significance (P ⬍ .15) were
allowed to enter the model in a stepwise fashion. Additional
adjustment was made for variations across study sites by
forcing in the model those institutions with univariable
significance for the outcome variable (P ⬍ .15). A similar
approach was followed in subsequent logistic regression
models to determine whether absence of pain itself was
independently associated with hospital mortality.
RESULTS
Of the 10,783 patients, ACS was diagnosed in 2,541
(24%). Of these, at final diagnosis assignment, 894 (35%)
were found to have AMI and 1,647 (65%) acute ischemia
without infarction.
Clinical Characteristics
There were clear baseline clinical differences among patients with an ACS between those with and without pain
(Table 1). Patients without pain at presentation were significantly older (71 years v 65 years, P ⫽ .0001). Additionally,
there was a higher proportion of women among patients
without pain (53%, P ⫽ .007), while the opposite occurred
in the group with pain present (42%, P ⫽ .007). Patients
with painless ACS had a higher prevalence of diabetes
mellitus (35% v 26%, P ⫽ .01) and had a trend for higher
rates of prior infarction by history (52% v 44%, P ⫽ .06).
Chief complaint data showed that chest pain was the most
frequent chief complaint among those patients with pain at
presentation (95%), and dyspnea was the most frequent
chief complaint in those with painless ischemia (72%).
Finally, patients with a painless ACS more commonly had
AMI than those with pain (9.8% v 4.2%, P ⫽ .001).
Table 2 shows that among patients with AMI, severe
degrees of congestive heart failure on admission were more
frequent among patients without pain, as evidenced by
Killip class III (36% v 8.5%, P ⫽ .001) and Killip class IV
(5.8% v 3.2%, P ⫽ .001). Patients with painless AMI had
similar delays from time of symptom onset to ED arrival as
patients with pain. However, among all AMI patients that
died in the hospital, significant longer delays in ED arrival
were seen in those with painless AMI (30 hours v12.5 hours,
P ⫽ .02).
TABLE 1. Characteristics of Patients with Cardiac Ischemia by
Clinical Presentation (n ⫽ 2,541)
Pain Absent
(n ⫽ 158)
Mean age (years)
Women (%)
Diabetes (%)
History of AMI (%)
History of hypertension (%)
Presenting symptoms (%)
Chest pain
Dyspnea
Syncope
Dizziness
Other
AMI (%)
Pain Present
(n ⫽ 2,379)
P value
71
53
35
52
64
65
42
26
44
62
.0001
.007
.01
.06
.6
—
72
8.9
8.2
11
9.8
95
4.7
0.04
0.04
0.7
4.2
.001
Abbreviation: AMI, acute myocardial infarction.
Emergency Department Triage
The effect of presenting symptoms on ED triage was
analyzed depending on the presence or absence of pain.
There were no overall differences in ED triage between the
two groups among patients with an ACS (overall ChiSquare P ⫽ .9 for admission to medical ward, monitored
bed, critical care bed or discharge to home). However, in the
subgroup of patients with AMI (n⫽894), significantly fewer
patients without pain were admitted to a critical care unit as
opposed to similar patients with pain (51% v 67%, P ⫽
.003).
Predicting Painless Ischemia
To determine factors that may be associated with painless
ACS, we constructed a logistic regression model with painless ischemia as the outcome variable (Table 3). The final
model (receiver-operating characteristic curve area 0.75)
identified age and congestive heart failure as independent
predictors of lack of pain among patients with cardiac
ischemia. A second logistic regression model performed in
a similar fashion for the subgroup of patients with AMI
identified diabetes mellitus, in addition to age and congestive heart failure (Table 4). Both models exhibited a statistically significant interaction between age and congestive
heart failure. This relationship is depicted in Figure 1. This
graph illustrates several points. First, the rates of “painlessness” increase gradually with age in patients without conTABLE 2. Characteristics of Patients with Acute Myocardial
Infarction by Clinical Presentation (n ⫽ 894)
Killip class (%)
I
II
III
IV
Median time
to ED
(hours)
Pain Absent
(n ⫽ 87)
Pain Present
(n ⫽ 807)
41 (48%)
17 (20%)
36 (41%)
5.8 (6.7%)
3.0 (0.25 to ⬎72)
71 (8.8%)
17 (2.1%)
8.5 (9.8%)
3.2 (4%)
2.8 (0.17 to ⬎72)
P value
.001
.5
CORONADO ET AL ■ ACS WITHOUT PAIN
TABLE 3.
571
Univariable and Multivariable Odds Ratios for Clinical Characteristics Associated with Painless Cardiac Ischemia (n ⫽ 2,541).
Univariable Analysis
Age †
Heart failure ‡
Diabetes
Female gender
Hypertension
Prior infarction
Multivariable Logistic Regression*
OR
95% CI
P value
OR
95%CI
P value
1.5
4.6
1.5
1.5
1.1
1.4
(1.3-1.7)
(3.3-6.6)
(1.0-2.0)
(1.1-2.1)
(0.8-1.6)
(1.0-2.0)
0.0001
.0001
.03
.008
.5
.04
1.6
51
—
—
—
—
(1.4-4.7)
(5.4-473)
—
—
—
—
.0001
.0006
ns
ns
ns
ns
*Multivariable model controlled for variations across study sites. Receiver-operating characteristic curve area 0.75.
†Odds ratio per decade of life.
‡Defined as radiographic evidence of pulmonary congestion and/or edema.
gestive heart failure upon ED arrival. Second, the presence
of congestive heart failure is associated with higher rates of
painless ischemia across age groups, and third, beyond age
40 the rate of painless ischemia in patients with heart failure
remains relatively constant.
Hospital Mortality
The unadjusted hospital mortality rates were higher
among patients with painless ischemia than in those with
pain, with similar differences among those with AMI (18%
v 6.9%). As noted earlier, patients with painless AMI had a
trend for higher Killip classes. However, no statistically
significant differences in hospital mortality were noted between patients with painful and painless AMI when comparisons were made within each Killip class (Class I, 15%
v 3.6%, P ⫽ .08; class II 13% v 11%, P ⫽ .8; class III 19%
v 7.4%, P ⫽ .1 and class IV 40% v 54%, P ⫽ .6, respectively).
To further determine the effect of painless ischemia on
hospital mortality, we constructed another logistic regression model among all patients with acute ischemia with
“lack of pain” as an independent variable entered in a
stepwise fashion (Table 5). Other variables known or suspected to be associated with hospital mortality from acute
cardiac ischemia were also allowed to enter the model in a
stepwise fashion. They included age, heart failure on presentation, diabetes mellitus, and blood pressure. Blood pressure was entered as a previously validated function with
linear and quadratic components.33 Additional statistical
adjustment in the model was made for institutional variation, hospital interventions (administration of thrombolytic
TABLE 4.
agents, cardiac catheterization, coronary angioplasty, coronary artery bypass surgery), and time from symptom onset
and triage from the ED. In addition to age, heart failure on
presentation, diabetes, and presenting blood pressure, lack
of pain was identified as an independent predictor of hospital mortality.
DISCUSSION
The clinical profile of patients with symptomatic ACS
can be best studied by including as wide a spectrum of
presenting symptoms as possible for patients seeking medical care. Capturing patients with a broad range of symptoms will not only allow the analysis of a more complete
clinical spectrum but also facilitates determining the effect
that clinical presentation itself has on medical outcomes.
This study illustrates several features with respect to the
presence or absence of pain in patients with an ACS. First,
there are clear differences in the demographics between the
2 groups. Patients with a painless ACS tended to be older,
were more frequently women, and more frequently had
diabetes and prior infarctions, findings that have also been
supported by other reports.11,26 Higher rates of painless
ACS in older patients and in women are consistent with the
rates of clinically unrecognized infarctions detected on routine electrocardiograms in patients from the Framingham
Study.5 Why women had higher rates of painless ischemia
than men is not readily apparent, but older age at presentation and higher prevalence of diabetes are likely to play a
role. Higher rate of prior infarction seen in patients with
painless ischemia is consistent with the suggested notion
Univariable and Multivariable Odds Ratios for Clinical Characteristics Associated with Painless Myocardial Infarction (n ⫽ 894).
Univariable Analysis
Age †
Heart failure ‡
Diabetes
Female gender
Hypertension
Prior infarction
Multivariable Logistic Regression*
OR
95% CI
P value
OR
95% CI
P value
1.8
3.6
1.9
2.0
1.4
1.2
(1.5-2.2)
(2.2-5.7)
(1.2-3.0)
(1.3-3.2)
(0.9-2.3)
(0.8-1.9)
0.0001
0.0001
0.005
0.002
0.12
0.4
2.0
79
1.6
—
—
—
(1.6-2.7)
(3.2-999)
(1.0-2.7)
—
—
—
0.0001
0.008
0.05
ns
ns
ns
*Multivariable model controlled for variations across study sites. Receiver-operating characteristic curve area 0.77.
†Odds ratio per decade of life.
‡Defined as radiographic evidence of pulmonary congestion and/or edema.
AMERICAN JOURNAL OF EMERGENCY MEDICINE ■ Volume 22, Number 7 ■ November 2004
572
FIGURE 1. Lack of pain in acute cardiac ischemia in patients
with and without congestive heart failure (CHF), n⫽2,541.
that prior myocardial damage impairs pain perception due to
disruption of sensory receptors.8,33
The absence of pain during episodes of acute cardiac
ischemia has been well documented for many years.4-12
Since early reports of pathologically documented painless
AMI in patients with diabetes,34 different mechanisms that
explain impaired pain perception in these patients have been
postulated. As suggested by other investigators, autonomic
neuropathy shown in hemodynamic studies in some patients
with diabetes may correlate with a pure cardiac sensory
neuropathy.33,37 Psychological and neuropsychiatric factors
have also been implicated as causes of impaired pain perception.35
Regardless of the precise mechanisms that explain the
presence or absence of pain during acute ischemia, clinical
presentation per se has received little attention in the research community as it relates to clinical outcomes. Our
lack of knowledge stems in part from selection bias, where
patients with atypical presentations, and more importantly,
those without chest pain, are systematically excluded from
many clinical trials. It is likely that atypical presentations
themselves lead to systematic poor recognition of ACSs by
investigators and clinicians due to their focus on the more
TABLE 5.
“classic” symptoms. Lastly, patient ignorance and even
denial can also play a role in under-reporting of symptoms
potentially due to an ACS.
Although epidemiological studies such as Framingham
Heart Study5 yielded useful information about clinically
unrecognized AMI in the community, they failed to provide
data on clinical presentation in acutely symptomatic patients.
Another factor that has perhaps limited our understanding
of the clinical manifestations of coronary disease is the
perception of AMI and unstable angina as separate entities.
Acute coronary syndromes are continuums that share a
common pathophysiology where the extent of myocardial
damage depends mainly on the duration of coronary occlusion, the presence of collateral circulation and the occurrence of spontaneous or pharmacologic reperfusion (27,28).
Studying only patients with documented AMI therefore
ignores a wealth of clinical information in patients also at
risk for complications and death.
Why diabetes was an independent predictor of painless
AMI but not among all patients with acute ischemia remains
unclear, but it is possible that the faulty pain perception seen
in these patients caused only those with severe ischemia and
subsequent infarction to seek medical care. It is possible that
those with more subtle symptoms and less severe ischemia
never sought medical care and were never enrolled in the
trial.
Although the rates of painless ischemia increased with
age, the presence of congestive heart failure upon arrival to
the ED conferred an excess risk for presenting without pain,
particularly in those over 40 years of age. This finding is
likely to indicate that congestive heart failure, rather than
causative, is an epiphenomenon in patients with blunted
pain perception who develop ischemia. In other words,
rather than ventricular failure predisposing patients to painless ischemia, those with ischemia and no pain will eventually present with signs of ventricular failure if their myocardial perfusion has been sufficiently compromised. In this
case, breathlessness and pulmonary edema become their
first and perhaps only “warning sign.”
Although this and other studies have shown higher mortality rates in patients who develop congestive heart failure,
factors other than a greater loss of myocardial mass seem to
be at play. Jaffe et al34 reported a correlation between heart
failure and excess mortality despite smaller infarct size in a
Unadjusted and Adjusted Odds Ratios for Hospital Mortality in Patients With Acute Cardiac Ischemia (n ⫽ 2,541).
Univariable Analysis
Age †
Heart failure ‡
Absence of pain
Diabetes
Blood pressure§
Multivariable Logistic Regression*
OR
95% CI
P value
OR
95% CI
P value
1.7
4.1
3.9
1.7
—
(1.4-2.1)
(2.6-6.4)
(2.2-6.9)
(1.1-2.6)
—
.0001
.0001
.0001
.02
.0001
1.8
2.9
4.3
2.2
—
(1.4-2.2)
(1.7-4.7)
(2.0-9.3)
(1.3-3.7)
—
.0001
.0001
.0002
.003
.01
*Multivariable model controlled for variations across study sites, hospital interventions (administration of a thrombolytic agent, cardiac
catheterization, coronary angioplasty, coronary bypass surgery), admission hospital ward, and time from symptom onset to hospital arrival. The
receiver-operating characteristic curve area was 0.85.
†Odds ratio per decade of life.
‡Defined as radiographic evidence of pulmonary congestion and/or edema.
§Reported as previously validated function with linear and quadratic components.
CORONADO ET AL ■ ACS WITHOUT PAIN
group of 100 diabetics. The reason for this is not completely
understood, but altered myocardial remodeling,36 preexisting cardiac dysfunction38 and more extensive coronary atherosclerosis have all been invoked. In patients with diabetes,
other investigators have described myocardial lesions that
do not correlate with coronary pathology,41 whereas others
have even described a distinct cardiomyopathy.37
It is important to note that painless cardiac ischemia had
an effect on patient triage by the emergency physician and
may have lowered the level of suspicion for cardiac ischemia. Although there was no overall difference in ED triage
among all patients with ischemia, significantly fewer patients with painless presentations were admitted to a critical
care bed among those with AMI. To what extent the excess
mortality seen in patients without pain can be attributed to
poor symptom recognition and to sub-optimal monitoring
and/or treatment remains to be elucidated and cannot be
determined from the data obtained in this study. However,
just as delays in ED arrival and proper intervention correlate
with poor outcomes, suboptimal treatment for ongoing undetected ischemia after hospital admission may have affected clinical outcomes in a similar fashion.40
This study has several limitations. There have been many
changes in the practice of emergency medicine, which have
improved the diagnosis and treatment of ACS since these
data were collected in 1993, yet the design and size of this
study make its findings applicable to this day. In fact, it is
likely that this study underestimated the magnitude of painless ischemia in our communities given our aging population and growing numbers of newly diagnosed diabetics.
Conversely, it is possible that some patients with AMI may
have gone undetected given the unavailability of testing for
serum troponins at the time of the trial. It is also possible
that despite broad inclusion criteria, physicians’ suspicion
for cardiac ischemia may have caused overenrollment of
certain patient groups, such as diabetics with atypical presentations. However, in this study AMI was identified
through the routine measurement of cardiac enzymes on
every patient enrolled in an effort to minimize ascertainment bias.
Another limitation of this study was the exclusion of rural
hospitals or hospitals with chest pain units (not in wide
spread use at the time of the trial) and of hospitals without
emergency physicians on site. However, similar findings
were observed in rural hospitals in our earlier study, which
used the same inclusion criteria and follow-up methods.42
Lastly, another possible limitation, the exclusion of patients
with a language barrier, may have led to an underrepresentation of certain ethnic groups. The high inclusion rate
(92%) makes this factor of unlikely significance.
CONCLUSIONS
The excess mortality rates seen in patients with painless
ACS can be at least partially explained by higher rates of
congestive heart failure at presentation, possibly in conjunction with a longer delay between time of onset of ischemia
to ED arrival and poor symptom recognition by both patient
and physician. The lack of significant differences in time
delay between patients with and without pain may reflect a
faulty warning signal that allows ischemia to go unnoticed
until there is a large ischemic burden that causes other
symptoms such as dyspnea to develop.
573
Further studies are needed to determine the exact nature
of painless ACS, to guide future efforts to improve the ED
diagnosis of ACS, to optimize patient triage and to heighten
patient awareness of the manifestations of ACS. Although
some studies have studied the use of continuous 12-lead
electrocardiographic monitoring in the ED for improved
diagnosis of cardiac ischemia and infarction,39 large prospective studies along with ongoing patient education will
likely provide the information necessary to ensure early
diagnosis and treatment in persons whose manifestations of
acute cardiac ischemia are of difficult recognition by both
patient and physician. We may need to find a better label
than “chest pain” to describe the warning symptoms of an
ACS if we expect the public to properly understand and
respond promptly to their symptoms.
REFERENCES
1. Heberden W: Commentaries on the history and cause of disease, Boston p 292, Wells & Lilly, 1818
2. Gillium RF, Fortmann SP, Prineas RJ, et al: International diagnostic criteria for acute myocardial infarction and acute stroke. Am
Heart J 1984;108:150-158
3. The Joint European Society of Cardiology/American College of
Cardiology Committee: Myocardial infarction redefined-A consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial infarction. J Am Coll Cardiol 2000;36:959-69
4. Bean WB: Masquerades of myocardial infarction. Lancet 1977;
1:1044-1046
5. Kannel WB, Abbott RD: Incidence and prognosis of unrecognized myocardial infarction. An update on the Framingham study
N Engl J Med 1984;311:1144-1147
6. Uretsky BF, Farquhar DS, Berezin AF, et al: Symptomatic
myocardial infarction without chest pain: prevalence and clinical
course. Am J Cardiol 1977;40:498-503
7. Turiel M, Galassi AR, Glazier JJ, et al: Pain threshold and
tolerance in women with syndrome X and women with stable angina
pectoris. Am J Cardiol 1987;60:503-8
8. Keele KD: Pain complaint threshold in relation to pain of cardiac infarction. Br Med J 1968;1:670-3
9. Pope JH, Ruthazer R, Beshansky JR, et al: The clinical presentation of patients with acute cardiac ischemia in the emergency
department: a multicenter controlled clinical trial. J Thromb Thrombolysis 1998;6:63-74
10. Milner KA, Funk M, Richards S, et al: Gender differences in
symptom presentation associated with coronary heart disease.
Am J Cardiol 1999;84:396-399
11. Canto JG, Shlipak MG, Rogers WJ, et al: Prevalence, clinical
characteristics, and mortality among patients with myocardial infarction presenting without chest pain. JAMA 2000;283:3223-3229
12. Umachandran V, Ranjadalayan K, Ambepityia G, et al: The
perception of angina in diabetes: relation to somatic pain threshold
and autonomic function. Am Heart J 1991;121:1649-54
13. Barron HV, Bowlby LJ, Breen T, et al: Use of reperfusion
therapy for acute myocardial infarction in the United States. Circulation 1998;97:1150-1156
14. Nadelmann J, Frishman WH, Ooi WL, et al: Prevalence, incidence and prognosis of recognized and unrecognized myocardial
infarction in persons aged 75 years or older: The Bronx Aging Study.
Am J Cardiol 1990;66:533-7
15. Acharya DU, Shekhar YC, Aggarwal A, et al: Lack of pain
during myocardial infarction in diabetics: is autonomic dysfunction
responsible? Am J Cardiol 1991;68:793-6
16. Rosenman RH, Friedman M, Jenkins CD, et al: Clinically
unrecognized myocardial infarction in the Western Collaborative
Group Study. Am J Cardiol 1967;19:776-82
17. Medalie JH, Gouldbourt U: Unrecognized myocardial infarction: five-year incidence, mortality, and risk factors. Ann Intern Med
1976;84:526-31
18. Sheifer SE, Manolio TA, Gersh BJ: Unrecognized myocardial
infarction. Ann Intern Med 2001;135:801-811
574
AMERICAN JOURNAL OF EMERGENCY MEDICINE ■ Volume 22, Number 7 ■ November 2004
19. Abbott RD, Donahue RD, Kannel WB: The impact of diabetes
on survival following myocardial infarction in men v women (The
Framingham Study): JAMA 1988;260:3456-60
20. Kouvaras G, Cokkinos D, Spyropoulou M: Increased mortality
of diabetics after acute myocardial infarction attributed to diffusely
impaired left ventricular performance as assessed by echocardiography. Jpn Heart J 1988;1:19
21. Soler NG, Bennet MA, Pentecost BL, et al: Myocardial infarction in diabetics. Q J Med 1975;44:125-32
22. Partamian JO, Bradley RF: Acute myocardial infarction in 258
cases of diabetes. Immediate mortality and five-year survival 1965;
273:455-61
23. Jacoby RM, Nesto RW: Acute myocardial infarction in the
diabetic patient: pathophysiology, clinical course and prognosis.
J Am Coll Cardiol 1992;20:736-44
24. Savage MP, Krolewski AS, Kenien GC, et al: Acute myocardial infarction in diabetes mellitus and significance of congestive
heart failure as a prognostic factor. Am J Cardiol 1988;62:665-9
25. Lee KL, Woodlief LH, Topol EJ, et al: Coronary artery disease/
myocardial infarction: Predictors of 30-day mortality in the era of
reperfusion for acute myocardial infarction: Results from an international trial of 41,021 patients. Circulation 1995;91:1659-68
26. Goldstein RE, Boccuzzi SJ, Cruess D, et al: Prognosis after
hospitalization for acute myocardial infarction not accompanied by
typical ischemic chest pain. Am J Med 1995;99:123-31
27. Chesebro JH, Foster V: Dynamic thrombosis and thrombolysis: role of antithrombins [editorial]. Circulation 1991;83:1815-7
28. Falk EA: Atherogenesis and thrombosis, in Califf RM, Mark
DB, Wagner GS (eds): Acute Coronary Care (ed2) St. Louis, MO,
Mosby-year Book, 1995, p 711
29. Selker HP, Beshansky JR, Griffith JL, et al: Use of the acute
cardiac ischemia time-insensitive predictive instrument (ACI-TIPI) to
assist with triage of patients with chest pain or other symptoms
suggestive of acute cardiac ischemia: A multicenter,controlled clinical trial. Ann Int Med 1998;129:845-55
30. Van der Does E, Lubson J, Pool J, et al: Acute coronary
events in a general practice: objectives and design of the Imminent
Myocardial Infarction Rotterdam Study. Heart Bull 1976;7:91-8
31. Killip T, Kimball JT: Treatment of myocardial infarction in a
coronary careunit: a two year experience with 250 patients. Am J
Cardiol 1967;20:457-64
32. Campeau L: Grading of angina pectoris. Circulation 1976;54:
522-523
33. Niakan E, Harati Y, Rolak LA, et al: Silent myocardial infarction and diabetic cardiovascular autonomic neuropathy. Arch intern
Med 1986;146:2229-2230
34. Jaffe AS, Spadaro JJ, Schechtman K, et al: Increased congestive heart failure after myocardial infarction of modest extent in
patients with diabetes mellitus. Am Heart J 1984;108:31-7
35. Torosian T, Lumley MA, Pickard SD, et al: Silent versus
symptomatic myocardial ischemia: the role of psychological and
medical factors. Health Psychol 1997;16:123-30
36. Iwasaka T, Takiahashi N, Nakamura S, et al: Residual left
ventricular pump function after acute myocardial infarction in
NIDDM patients. Diab Care 1992;15:1522-6
37. Galderisi M, Anderson KM, Wilson PWF, et al: Echocardiographic evidence for the existence of a distinct diabetic cardiomyopathy (The Framingham Heart Study). Am J Cardiol 1991;68:85-9
38. Lomuscio A, Bestetti A, Vergani D, et al: Radionuclide assessment of left ventricular function in patients with myocardial
infarction and diabetes mellitus. J Intern Med 1992;231:73-6
39. Fesmire FM, Percy RF, Bardoner JB, et al: Usefulness of
automated serial 12-lead ECG monitoring during the initial emergency department evaluation of patients with chest pain. Ann Emerg
Med 1998;31:3-11
40. Pope JH, Aufderheide T, Ruthazer R, et al: Missed diagnoses
of acute cardiac ischemia in the emergency department. N Engl
J Med 2000;342:1163-70
41. Burchfiel CM, Reed DM, Marcus EB, et al: Association of
diabetes mellitus with coronary atherosclerosis and myocardial lesions: An autopsy study from the Honolulu Heart program. Am J
Epidemiol 1993;137:1328-40
42. Larsen GC, Griffith JL, Beshansky JR, et al: Electrocardiographic left ventricular hypertrophy in patients with suspected acute
coronary syndrome-Its influence on diagnosis, triage, and shortterm prognosis: A multicenter study. J Gen Intern Med 1994;9:666673