6. Initial Recognition And
Management In The Emergency Department
A variety of treatment options (Figure
3) (24-40)
are available that can reduce mortality and morbidity in patients
with STEMI, but the effectiveness of these therapies diminishes
rapidly within the first several hours after symptoms onset (162,185).
The traditional ED evaluation of patients with chest pain relies
heavily on the patient’s history, physical examination, and
the ECG. This approach not infrequently fails to identify patients
who are actually suffering from STEMI, which results in an inappropriate
discharge home from the ED (186).
Such missed MI patients are at relatively high risk of death or
complications for the next 4 to 6 weeks after ED discharge (187-192).
In a large study on this subject, Pope et al. (188)
found that 889 of 10 689 patients who presented to 10 US hospital
EDs with chest pain or other symptoms that suggested acute cardiac
ischemia had STEMI; 19 patients (2.1%, 95% CI 1.1% to 3.1%) were
discharged from the ED. Patients with STEMI were more likely not
to be hospitalized if they were nonwhite (odds ratio [OR] for discharge
4.5; 95% CI 1.8 to 11.8) or had a normal or nondiagnostic ECG (OR
7.7; 95% CI 2.9 to 20.2). The risk-adjusted mortality ratio for
MI patients who were not hospitalized compared with those who were
hospitalized was 1.9 (95% CI 0.7 to 5.2).
6.1.
Optimal Strategies for Emergency Department Triage
Class I
Hospitals should establish multidisciplinary teams (including
primary care physicians, emergency medicine physicians, cardiologists,
nurses, and laboratorians) to develop guideline-based, institution-specific
written protocols for triaging and managing patients who are seen
in the prehospital setting or present to the ED with symptoms suggestive
of STEMI. (Level of Evidence: B)
The advent of highly effective, time-dependent treatment for STEMI
coupled with the need to reduce healthcare costs adds further incentive
for clinicians to get the right answer quickly and to reduce unnecessary
admissions and length of hospital stay. Investigators have tried
various diagnostic tools such as clinical decision algorithms, cardiac
biomarkers, echocardiography, and myocardial perfusion imaging in
an attempt to avoid missing patients with MI or unstable angina.
The most successful strategies to emerge thus far are designed to
identify MI patients and, when clinically appropriate, screen for
unstable angina and underlying coronary artery disease. Most strategies
use a combination of cardiac biomarkers, short-term observation,
diagnostic imaging, and provocative stress testing. An increasing
number of high-quality centers now use structured protocols, checklists,
or critical pathways to screen patients with suspected MI or unstable
angina (193-205).
It does not appear to matter whether the institution designates
itself a chest pain center. Rather, it is the multifaceted, structured
approach to the problem that appears to provide clinical, cost-effective
benefit (206,207).
One randomized trial has confirmed the safety, efficacy, and cost-effectiveness
of the structured decisionmaking approach compared with standard,
unstructured care (208).
6.2. Initial Patient Evaluation
Class I
1. The delay from patient contact with the healthcare system (arrival
at the ED or contact with paramedics) to initiation of fibrinolytic
therapy should be less than 30 minutes. Alternatively, if PCI is
chosen, the delay from patient contact with the healthcare system
(typically, arrival at the ED, or contact with paramedics) to balloon
inflation should be less than 90 minutes. (Level of Evidence: B)
2. The choice of initial STEMI treatment should be made by the emergency
medicine physician on duty based on a predetermined, institution-specific,
written protocol that is a collaborative effort of cardiologists
(both those involved in coronary care unit management and interventionalists),
emergency physicians, primary care physicians, nurses, and other
appropriate personnel. For cases in which the initial diagnosis
and treatment plan are unclear to the emergency physician
or are not covered directly by the agreedupon protocol, immediate
cardiology consultation is advisable. (Level of Evidence: C)
Regardless of the approach used, all patients presenting to the
ED with chest discomfort or other symptoms suggestive of STEMI or
unstable angina should be considered high-priority triage cases
and should be evaluated and treated on the basis of a predetermined,
institution-specific chest pain protocol. The protocol should include
several diagnostic possibilities (Figure
9) (4). The patient should be
placed on a cardiac monitor immediately, with emergency resuscitation
equipment, including a defibrillator, nearby. An ECG should be performed
and shown to an experienced emergency medicine physician within
10 minutes of ED arrival. If STEMI is present, the decision as to
whether the patient will be treated with fibrinolytic therapy or
primary PCI should be made within the next 10 minutes (Figure
7) (180). The goal for patients
with STEMI should be to achieve a door-to-needle time of within
30 minutes and a door-to-balloon time of within 90 minutes (Figure
6) (155). If the initial ECG
is not diagnostic, the patient remains symptomatic, and there is
a high clinical suspicion for STEMI, serial ECGs at 5- to 10-minute
intervals or continuous ST-segment monitoring should be performed.
Ideally, such decisions should be made by the emergency medicine
physician on duty in the ED based on a predetermined, institution-specific,
written protocol that has been developed with input from cardiologists
(both those involved in coronary care unit management and interventionalists),
emergency medicine physicians, primary care physicians, nurses,
and other appropriate personnel. For non-interventional hospitals,
this will usually require formal, written transfer agreements and
protocols that will permit expeditious transfer of patients who
require urgent mechanical revascularization
to the nearest appropriate interventional facility (Figure
6) (155). The protocol should
also include the level of training and certification of personnel
required to accompany the patient during transfer, the minimum equipment
requirements, and the type(s) of transport vehicles (e.g., standard
ground ambulance, mobile intensive care unit, helicopter, or fixed-wing
aircraft) that can be used on the basis of patient condition. For
cases in which the initial diagnosis and treatment plan are unclear
to the emergency medicine physician or are not covered directly
by the agreed-upon protocol, immediate cardiology consultation is
advisable.
6.2.1. History
Class I
The targeted history of STEMI patients taken in the ED should ascertain
whether the patient has had prior episodes of myocardial ischemia,
such as stable or unstable angina, MI, coronary bypass surgery,
or PCI. Evaluation of the patient’s complaints should focus
on chest discomfort, associated symptoms, sex and age-related differences
in presentation, hypertension, diabetes mellitus, possibility of
aortic dissection, risk of bleeding, and clinical cerebrovascular
disease (amaurosis fugax, face/limb weakness or clumsiness, face/limb
numbness or sensory loss, ataxia, or vertigo). (Level of Evidence:
C)
The history taken in the ED must be concise and detailed enough
to establish the probability of STEMI but should be obtained expeditiously
so as not to delay implementation of reperfusion therapy.
Chest Discomfort. The severity of discomfort varies and
is typically graded on a scale of 1 to 10, with 10 being the mostsevere
pain. It is important to keep in mind that many patients will not
admit having chest “pain” but will acknowledge the presence
of chest “discomfort,” because of their definition of
pain. The chest discomfort is often described as a crushing, vice-like
constriction, a feeling equivalent to an “elephant sitting
on the chest,” or heartburn. Usually, the discomfort is substernal
but may originate in or radiate to areas such as the neck, jaw,
interscapular area, upper extremities, and epigastrium. The duration
of the discomfort, which typically lasts longer than 30 minutes,
may wax and wane and may be remitting. It may be described as “indigestion
in the chest” and occasionally may be relieved with belching.
The possibility of precipitation of STEMI by use of illicit drugs
such as cocaine should be considered.
The targeted history of patients with STEMI taken in the ED should
ascertain whether the patient has had prior episodes of myocardial
ischemia such as stable or unstable angina, MI, coronary bypass
surgery, or PCI. Evaluation of the patient’s complaints should
focus on chest discomfort, associated symptoms, sex- and age-related
differences in presentation, hypertension, diabetes mellitus, possibility
of aortic dissection, risk of bleeding, and clinical cerebrovascular
disease (amaurosis fugax, face/limb weakness or clumsiness, face/limb
numbness or sensory loss, ataxia, or vertigo).
Associated Symptoms. Other symptoms to be aware of when
taking a patient’s history include nausea and vomiting. Diaphoresis
associated with a pale complexion may also appear, as well as weakness
or profound fatigue. Dizziness, lightheadedness, syncope, and paresthesia
may occur because of pain and hyperventilation.
Hypertension. Hypertension should be assessed, because chronic,
severe, poorly controlled hypertension or severe uncontrolled hypertension
on presentation is a relative contraindication to fibrinolytic therapy
(see Section 6.3.1.6.3.2).
Sex-
and Age-Related Differences in Presentation. It has been noted
in studies that women present with STEMI at an older age and later
after the onset of symptoms than men (53,210).
There must be an elevated index of suspicion during the evaluation
of women for STEMI. Although some variation exists, when large databases
of MI patients are examined, symptom profiles for STEMI by sex generally
appear more similar than different between men and women (211-215).
Elderly patients with STEMI are significantly less likely than younger
patients to complain of chest discomfort. However, elderly patients
with STEMI are more likely to complain of shortness of breath, as
well as other atypical symptoms such as syncope or unexplained nausea.
(181).
Diabetes Mellitus. Diabetics may have impaired angina (pain)
recognition, especially in the presence of autonomic neuropathy.
A diabetic may misinterpret dyspnea, nausea, vomiting, fatigue,
and diaphoresis as disturbance of diabetic control. Up to 50% of
diabetic individuals with type 2 diabetes for longer than 10 years
will have autonomic nervous system dysfunction manifested by impaired
heart rate variability.
Diabetics with STEMI should be evaluated for renal dysfunction (216).
Possibility of Aortic Dissection. Severe tearing pain radiating
directly to the back associated with dyspnea or syncope and without
ECG changes indicative of STEMI should raise the suspicion for aortic
dissection, and appropriate studies should be undertaken. Clinicians
should have a heightened index of suspicion for aortic dissection
in elderly hypertensive patients. However, it must be kept in mind
that the dissection may extend to the pericardial sac and produce
cardiac tamponade or disrupt the origin of a coronary artery.
Risk of Bleeding. Patients should be questioned about
previous bleeding problems, e.g., during surgery or dental procedures,
history of ulcer disease, cerebral vascular accidents, unexplained
anemia, or melena. The use of antiplatelet, antithrombin, and fibrinolytic
agents as part of the treatment for STEMI will exacerbate any underlying
bleeding risks.
Clinical Cerebrovascular Disease. The patient with STEMI
frequently has medical conditions that are risk factors for both
MI and stroke. Evidence for prior episodes suggestive of clinical
cerebrovascular disease should be sought. For example, the patient
should be asked whether he/she has ever had symptoms of transient
retinal or cerebral ischemia such as amaurosis fugax, face/limb
weakness or clumsiness, face/limb numbness or sensory loss, ataxia,
or vertigo. Transient ischemic attacks (TIAs) typically last less
than 30 minutes, whereas symptoms that last more than 60 to 90 minutes
are more likely to indicate the presence of a stroke (217).
In addition, the patient should be asked whether he/she has ever
had an ischemic stroke, intracerebral hemorrhage [ICH], or subarachnoid
hemorrhage. A brief summary of the details for diagnosis of the
different stroke subtypes is available (218).
Finally, a history of cognitive decline/dementia may indicate the
presence of cerebral amyloid angiopathy and increased risk of ICH,
and information regarding head and facial trauma should be obtained.
6.2.2. Physical Examination
Class I
1. A physical examination should be performed to aid in
the diagnosis and assessment of the extent, location, and presence
of complications of STEMI. (Level of Evidence: C)
2. A brief, focused, and limited neurological examination to look
for evidence of prior stroke or cognitive deficits should be performed
on STEMI patients before administration of fibrinolytic therapy.
(Level of Evidence: C)
A brief physical examination may promote rapid triage (Table
5), whereas a more detailed physical examination aids in the
differential diagnosis and is useful for assessing the extent, location,
and presence of complications of STEMI (Tables
6 and 7) (219).
Evidence
of prior stroke or dementia may be suggested by the finding on examination
of focal neurological or cognitive deficits (Table
6). A brief but focused examination can identify focal neurological
or cognitive deficits.
6.2.2.1. Differential Diagnosis
The differential diagnosis of STEMI includes conditions that can
be exacerbated by fibrinolysis and anticoagulation (Table
8). The pain of aortic dissection is typically described as
searing, ripping, or tearing and frequently radiates to the back
or lower extremities. The pain is worse at onset and lasts for hours.
Major pulses may be absent, and a murmur of aortic regurgitation
may be present. A transesophageal echocardiogram, computed tomography
(CT) scan, or magnetic resonance imaging scan is useful for establishing
the diagnosis of aortic dissection. Active peptic ulcer disease
can be present with chest or epigastric pain, sometimes radiating
posteriorly, and may be associated with syncope, hematemesis, or
melena. Free subdiaphragmatic air may be seen on upright chest X-ray
in perforations. Acute pericarditis may show PR-segment depression
and ST-segment elevation on the ECG but without reciprocal ST-segment
depression (220). Pain from pericarditis
is usually pleuritic and can radiate to the shoulder and trapezius
ridge and is often relieved by sitting up and leaning forward, characteristics
not found in STEMI. A rub is often present. Pulmonary embolus, with
or without infarction, presents with dyspnea and knifelike pleuritic
pain, sometimes with hemoptysis. Pulmonary embolism can present
with chest pain similar to that of STEMI. Costochondral pain is
described as sharp or sticking, with associated localized tenderness.
Pneumothorax may present with acute dyspnea, pleuritic pain, and
differential decrease in breath sounds with hyperresonance over
1 lung field. Acute cholecystitis may mimic STEMI, and rightupper-
quadrant abdominal tenderness should be sought on physical examination.
6.2.3. Electrocardiogram
Class I
1. A 12-lead ECG should be performed and shown to an experienced
emergency physician within 10 minutes of ED arrival on all patients
with chest discomfort (or anginal
equivalent) or other symptoms suggestive of STEMI. (Level of
Evidence: C)
2. If the initial ECG is not diagnostic of STEMI but the patient
remains symptomatic, and there is a high clinical suspicion for
STEMI, serial ECGs at 5- to 10- minute intervals or continuous 12-lead
ST-segment monitoring should be performed to detect the potential
development of ST elevation. (Level of Evidence: C)
3. In patients with inferior STEMI, right-sided ECG leads should
be obtained to screen for ST elevation suggestive of RV infarction.
(See Section 7.6.6
and the ACC/AHA/ASE 2003 Guideline Update for the Clinical Application
of Echocardiography). (Level of Evidence: B)
The 12-lead ECG in the ED is at the center of the therapeutic decision
pathway because of the strong evidence that ST-segment elevation
identifies patients who benefit from reperfusion therapy (221).
Mortality increases with the number of ECG leads showing ST elevation.
Important predictors of mortality on the initial 12-lead ECG include
left bundle branch block (LBBB) and anterior location of infarction
(Figure 10) (222,223).
Diagnostic criteria of greater than 0.1 mV in leads V1 through V4
may have reduced specificity for STEMI in patients with early repolarization.
Some evidence exists to support the use of greater than or equal
to 0.2 mV anteroseptal elevation as a preferable threshold for diagnosing
STEMI, because a higher proportion of patients are correctly classified
as having STEMI than with a threshold of greater than 0.1 mV in
these leads (221).
In
the absence of ST elevation, there is no evidence of benefit of
fibrinolytic therapy for patients with normal ECG or nonspecific
changes, and there is some suggestion of harm (including increased
bleeding risk) for patients with ST-segment depression only (221,224).
Notwithstanding this, fibrinolytic therapy may be appropriate when
there is marked STsegment depression confined to leads V1 through
V4 and accompanied by tall R waves in the right precordial leads
and upright T waves indicative of a true posterior injury current
and circumflex coronary occlusion. In circumstances where there
is a suggestive clinical history and suggestive evidence of true
posterior infarction, confirmatory data from posterior leads (i.e.,
V7 and V8) as well as 2-dimensional echocardiography may be especially
helpful; this latter investigation has a high negative predictive
value (225,226).
Primary PCI is another reperfusion strategy that may be effective
in patients with true posterior MI (see Section
6.3.1.6.4.2).
Initial errors in ECG interpretation can result in up to 12% of
patients being categorized inappropriately, demonstrating a potential
benefit of accurate computer-interpreted electrocardiography and
fax transmission to an expert (227).
It is less likely that STEMI is present if the upward-directed STsegment
changes are concave rather than convex (228).
Because lethal ventricular arrhythmias may develop abruptly in patients
with STEMI, all patients should be monitored electrocardiographically
on arrival in the ED. It is important to examine serial tracings
approximately 5 to 10 minutes apart, or if symptoms recur, during
evaluation in the ED for development of ST elevation if the initial
ECG is nondiagnostic. ST elevation may also be detected by intermittent
visual inspection of the oscilloscope or auditory alarms in systems
with continuous ST-segment monitoring capability.
Although the Fibrinolytic Therapy Trialists’ (FTT) Collaborative
Group overview indicates that patients with new or presumably new
LBBB are at high risk when presenting with presumed MI, this ECG
presentation is a frequent cause of delay or lack of reperfusion
therapy because of the concern of the validity of the ECG criteria
for MI diagnosis and the risk of therapy. This is also a situation
in which direct PCI may be preferable to fibrinolytic therapy (156).
It has been suggested that patients with new or presumably new LBBB
coupled with a typical ischemic history be approached with a plan
to rule in MI using 1 of 3 ECG criteria that provide independent
diagnostic value. These consist of ST elevation greater than or
equal to 0.1 mV in leads with a positive QRS, ST depression greater
than or equal to 0.1 mV in V1 to V3, and ST elevation greater than
or equal to 0.5 mV in leads with a negative QRS (229,230).
6.2.4. Laboratory Examinations
Class I
Laboratory examinations should be performed as part of the management
of STEMI patients but should not delay the implementation of reperfusion
therapy. For specific laboratory examinations, see Table
9. (Level of Evidence: C)
In
addition to serum cardiac biomarkers for cardiac damage, several
routine evaluations have important implications for management of
patients with STEMI (Table 9). Although
these studies should be undertaken when the patient is first examined,
therapeutic decisions should not be delayed until results are obtained
because of the crucial role of time to therapy in STEMI.
6.2.5. Biomarkers of Cardiac Damage
Class I
1. Cardiac-specific troponins should be used as the optimum biomarkers
for the evaluation of patients with STEMI who have coexistent skeletal
muscle injury. (Level of Evidence: C)
2.
For patients with ST elevation on the 12-lead ECG and symptoms of
STEMI, reperfusion therapy should be initiated as soon as possible
and is not contingent on a biomarker assay. (Level of Evidence:
C)
Class
IIa
Serial biomarker measurements can be useful to provide supportive
noninvasive evidence of reperfusion of the infarct artery after
fibrinolytic therapy in patients not undergoing angiography within
the first 24 hours after fibrinolytic therapy. (Level of Evidence:
B)
Class III
Serial biomarker measurements should not be relied upon to diagnose
reinfarction within the first 18 hours after the onset of STEMI.
(Level of Evidence: C)
The nomenclature of acute coronary syndromes is illustrated
in Figure 2 (8-10).
The central position of the 12-lead ECG and initial triage of patients
are emphasized. Serum cardiac biomarkers (creatine kinase [CK],
CK-MB, cardiac-specific troponins, myoglobin) are useful for confirming
the diagnosis of MI and estimating infarct size. Serum cardiac biomarkers
also provide valuable prognostic informatin. For patients with ST-segment
elevation, the diagnosis of STEMI is secure; initiation of reperfusion
therapy should not be delayed while awaiting the results of a cardiac
biomarker assay (231,232)
(Table 10). Quantitative analysis
of cardiac biomarker measurements provides prognostic information
as well as a noninvasive assessment of the likelihood that the patient
has undergone successful reperfusion when fibrinolytic therapy is
administered (Figure 11) (233,234).
Because
there are differences in the clinical need for biomarke in STEMI
versus NSTEMI patients and differences in the characteristics of
the various cardiac biomarkers, preferential use of a particular
biomarker should be base on the clinical syndrome. CK-MB is found
in the skeletal muscle and blood of healthy subjects; therefore,
the cutoff value for an elevated CK-MB is typically set a few units
above the upper end of the reference (normal) range. In contrast,
because cardiac troponin I (cTnI) and cardiac troponin T (cTnT)
are not normally detected in the blood of healthy people, the definition
of an abnormally increased level is a value that exceeds that of
99% of a reference control group. Given the nearly absolute myocardial
tissue specificity and high sensitivity for even microscopic zones
of myocardial necrosis, the ACC and the European Society of Cardiology
subsequently declared cardiac troponins to be the preferred biomarker
for diagnosing MI (233). A single
cutoff point was recommended such that an MI would be diagnosed
if, as a result of myocardial ischemia, cTnI or cTnT were detected
at least once within 24 hours of the index clinical event at a level
exceeding the 99th percentile of the mean value measured in a normal
control population (233). The superior
sensitivity makes troponin the preferred marker for patients with
UA/NSTEMI. In contrast, patients with STEMI are recognized on the
basis of the 12-lead ECG, and in general, subsequent confirmation
of MI can be ascertained by measurement of any of the available
cardiac biomarkers. Occasionally, a very small infarct will be missed
by CK-MB; therefore, troponin should be measured for patients suspected
to have STEMI who have negative serial CK-MBs.
It
should be recognized that in patients with STEMI, cTnT and cTnI
may first begin to rise above the reference limit by 3 to 6 hours
from the onset of ischemic symptoms. Therefore, a significant number
of patients will present o the emergency room with negative biomarkers.
Myoglobin, a low-molecular-weight heme protein found in cardiac
and skeletal muscle, is not cardiac specific but is released more
rapidly from infarcted myocardium than CK-MB and may be detected
as early as 2 hours after STEMI.
In
some patients, cardiac-specific troponins may not be detectable
for up to 6 hours after onset of chest pain. Thus, when CK-MB, cTnI,
or cTnT levels are elevated in less than 6 hours after the onset
of discomfort in patients with STEMI, clinicians should suspect
that an antecedent episode of unstable angina was in fact MI and
the patient is exhibiting a stuttering course of occlusion and release
of the infarct artery. Data from the Global Utilization of Streptokinase
and TPA for Occluded Arteries (GUSTO) III Study suggest that patients
with STEMI who have elevated cTnT levels and who are less than 6
hours from the onset of discomfort have an increased mortality risk
(235).
CK-MB is the preferred, widely available cardiac biomarker for most
patients with STEMI, for whom the need to diagnose reinfarction
and noninvasively assess reperfusion is greater
than the need to make the diagnosis. By mapping the time course
of the rise and fall of a biomarker (typically CKMB), clinicians
can detect an interruption of the progressive fall of the biomarker
level to a point below the upper reference limit (Figure
11) (233,234).
Re-elevation of the biomarker level is evidence of myocardial reinfarction
(Figure 12). A more rapidly rising
and falling biomarker such as CKMB or myoglobin is superior for
diagnosing reinfarction. As a consequence of continuous release
from a degenerating contractile apparatus in necrotic myocytes,
elevations of cTnI may persist for 7 to 10 days after MI, and elevations
of cTnT may persist for up to 10 to 14 days. The more protracted
time course of kinetic release of cTnI and cTnT limits the ability
of clinicians to make the diagnosis of reinfarction within several
days after the index STEMI event. An algorithm illustrating the
decision making process that incorporates biomarker measurements,
ECG findings, clinical symptoms, and, if available, autopsy data
for making the diagnosis of reinfarction is shown in Figure
12.
In
addition to monitoring the patient for resolution of schemic-type
chest discomfort and regression of the magnitude of ST-segment elevation
on the ECG, clinicians can obtain serial measurements of serum cardiac
markers to buttress the noninvasive diagnosis of reperfusion of
the infarctrelated artery after fibrinolytic therapy (Figure
11) (233,234,236).
An early peak of CK-MB (12 to 18 hours) suggests reperfusion. Because
of its rapid-release kinetics, myoglobin is also an attractive marker
for the early diagnosis of reperfusion.
CK-MB
isoforms are another serum cardiac biomarker less frequently used
for evaluating patients with STEMI. CK-MB exists in only 1 form
in myocardial tissue but in different isoforms (or subforms) in
the plasma. An absolute level of CKMB2 greater than 1 U/L or a ratio
of CK-MB2 to CK-MB1 of 1.5 has improved sensitivity and specificity
for diagnosis of MI within the first 6 hours compared with conventional
assays for CK-MB (237).
6.2.5.1.
Bedside Testing for Serum Cardiac Biomarkers
Class I
1.
Although handheld bedside (point-of-care) assays may be used for
a qualitative assessment of the presence of an elevated level of
a serum cardiac biomarker, subsequent measurements of cardiac biomarker
levels should be done with a quantitative test. (Level of Evidence:
B)
2. For patients with ST elevation on the 12-lead ECG and symptoms
of STEMI, reperfusion therapy should be initiated as soon as possible
and is not contingent on a bedside biomarker assay. (Level of Evidence:
C)
Handheld rapid bedside assays are clinically available for measuring
cTnI, cTnT, myoglobin, and CK-MB, but in general, bedside assays
are less sensitive and less precise than quantitative assays. Small
desktop rapid analyzers are also available for the same purpose.
A rapid, high-voltage electrophoretic system is available for measuring
CK-MB isoforms. Monitoring the timing of the appearance of a positive
bedside assay result may provide clinicians with a tool for a semiquantitative
estimate of a serum cardiac biomarker level at the patient’s
bedside (238). A positive bedside
test should be confirmed by a conventional quantitative test. However,
reperfusion therapy should not be delayed while one awaits the results
of a quantitative assay.
6.2.6. Imaging
Class I
1. Patients with STEMI should have a portable chest X-ray, but this
should not delay implementation of reperfusion therapy (unless a
potential contraindication is suspected, such as aortic dissection).
(Level of Evidence: C)
2. Imaging studies such as a high-quality portable cest X-ray, transthoracic
and/or transesophageal echocardiography, and a contrast chest CT
scan or magnetic resonance imaging scan should be used for differentiating
STEMI from aortic dissection in patients for whom this distinction
is initially unclear. (Level of Evidence: B)
Class IIa
Portable echocardiography is reasonable to clarify the diagnosis
of STEMI and allow risk stratification of patients with chest pain
who present to the ED, especially if the diagnosis of STEMI is confounded
by LBBB or pacing or if there is suspicion of posterior STEMI with
anterior ST depressions. (See Section
7.6.7, Mechanical Causes of Heart Failure/Low-Output Syndrome.)
(Level of Evidence: B)
Class III
Single-photon emission CT (SPECT) radionuclide imaging should not
be performed to diagnose STEMI in
patients for whom the diagnosis of STEMI is evident on the ECG.
(Level of Evidence: B)
Various
forms of imaging are often used to evaluate patients with symptoms
that are suggestive of MI or acute coronary syndrome. Cardiac imaging
can be of value in further determining the cause of chest discomfort
in patients suspected of having an acute MI or unstable angina but
whose initial ECG is normal or nondiagnostic. The 2 most studied
techniques thus far have been echocardiography and radionuclide
imaging.
Bedside
echocardiography is useful for diagnosis and risk stratification
of chest pain patients in the ED (226).
A highquality portable chest X-ray, transthoracic and/or transesophageal
echocardiography, and a contrast chest CT scan can be useful for
differentiating acute MI from aortic dissection in patients for
whom this distinction is clinically unclear. SPECT radionuclide
imaging at rest is not routinely indicated to establish the diagnosis
of MI in patients with STEMI, although it can provide valuable,
accurate diagnostic and prognostic information in patients who present
to the ED with symptoms suggestive of acute cardiac ischemia and
a normal or nondiagnostic ECG (239).
During the recuperative phase of hospitalization for STEMI, SPECT
imaging can be used to study myocardial perfusion and to look for
segmental abnormalities of LV wall motion.
6.2.7. Global Risk Assessment Tools
Global risk assessment provides an opportunity to integrate various
patient characteristics into a single score that can convey an overall
estimate of a patient’s prognosis over a given period of time.
Beyond being informative about prognosis, the general value of these
risk assessment tools is that they can influence clinical strategies.
In general, the risk of the intervention should be commensurate
with the underlying risk of the patient without the intervention
and the expected benefit of the intervention. That is, a high-risk
intervention should usually not be used for a very low-risk patient.
The expected increase in risk associated with the intervention would
very likely outweigh the expected benefit.
Several
risk assessment tools have been proposed for patients with STEMI
(240-243).
One such tool uses clinical and ECG characteristics to predict risk
of mortality for a patient if and if not treated with fibrinolytic
therapy, as well as the risk of intracranial hemorrhage and major
bleeding. This decision aid suggests that some patients with small
infarctions may not have a substantial benefit from fibrinolytic
therapy, particularly those who may have a risk factor for bleeding.
These estimates are based on trials and registries. The use of this
aid in clinical practice did not increase the use of fibrinolytic
therapy overall (244). Whether
the widespread application of these tools can improve decision making
is not clear. Nevertheless, they provide estimates of risk that
may be useful in the tailoring of therapy for individual patients.
In general, however, patients who present with STEMI require evaluation
for rapid reperfusion therapy and treatment with aspirin, beta-blockers,
and ACE inhibitors. Nevertheless, any patient with a risk from the
intervention that exceeds their STEMI risk reduction will, on average,
do better without that treatment. This group will generally include
patients with a higher risk from the intervention or a lower absolute
risk reduction (generally because of a low absolute STEMI risk).
This issue may be particularly important for younger patients, who
tend to have a lower absolute risk of mortality (245),
and for the elderly, who tend to have a higher risk from interventions,
particularly with respect to fibrinolytic therapy (246).
Precise estimates of risks and benefits are useful because the low
STEMI risk in younger patients is often accompanied by a lower risk
of interventions. In contrast, in the elderly, the higher intervention
risk is accompanied by a higher STEMI risk (and thus a larger absolute
reduction in risk with the intervention) (247).
The
use of any risk assessment tool should not contribute to any delay
in providing the time-sensitive assessment and treatment strategies
that patients with STEMI require. Further research is necessary
to determine how these tools may best contribute to optimizing patient
outcomes.
6.3. Management
6.3.1. Routine Measures
6.3.1.1. Oxygen
Class I
Supplemental oxygen should be administered to patients with arterial
oxygen desaturation (SaO2 less than 90%). (Level of Evidence:
B)
Class IIa
It is reasonable to administer supplemental oxygen to all patients
with uncomplicated STEMI during the first 6 hours. (Level of
Evidence: C)
It has become universal practice to administer oxygen, usually by
nasal prongs, to virtually all patients suspected of having acute
ischemic-type chest discomfort, although it is not known whether
this therapy limits myocardial damage or reduces morbidity or mortality.
If oxygen saturation monitoring is used, therapy with supplemental
oxygen is indicated if the saturation is less than 90%. Experimental
results indicate that breathing oxygen may limit ischemic myocardial
injury (248), and there is evidence
that oxygen administration reduces ST-segment elevation (249).
The rationale for use of oxygen is based on the observation that
even with uncomplicated MI, some patients are modestly hypoxemic
initially, presumably because of ventilation-perfusion mismatch
and excessive lung water (250).
In
patients with severe congestive heart failure, pulmonary edema,
or a mechanical complication of STEMI, significant hypoxemia may
not be corrected with supplemental oxygen alone. Continuous positive-pressure
breathing or endotracheal intubation and mechanical ventilation
may be required in such cases (251).
For
patients without complications, excess administration of oxygen
can lead to systemic vasoconstriction, and high flow rates can be
harmful to patients with chronic obstructive airway disease. In
the absence of compelling evidence for established benefit in uncomplicated
cases, and in view of its expense, there appears to be little justification
for continuing its routine use beyond 6 hours.
6.3.1.2. Nitroglycerin
Class I
1. Patients with ongoing ischemic discomfort should receive sublingual
nitroglycerin (0.4 mg) every 5 minutes for a total of 3 doses, after
which an assessment should be made about the need for intravenous
nitroglycerin. (Level of Evidence: C)
2. Intravenous nitroglycerin is indicated for relief of ongoing
ischemic discomfort, control of hypertension, or management of pulmonary
congestion. (Level of Evidence: C)
Class III
1. Nitrates should not be administered to patients with systolic
blood pressure less than 90 mm Hg or greater than or equal to 30
mm Hg below baseline, severe bradycardia (less than 50 beats per
minute [bpm]), tachycardia (more than 100 bpm), or suspected RV
infarction. (Level of Evidence: C)
2. Nitrates should not be administered to patients who have received
a phosphodiesterase inhibitor for erectile dysfunction within the
last 24 hours (48 hours for tadalafil). (Level of Evidence:
B)
The physiological effects of nitrates include reducing preload and
afterload through peripheral arterial and venous dilation, relaxation
of epicardial coronary arteries to improve coronary flow, and dilation
of collateral vessels, potentially creating a more favorable subendocardial
to epicardial flow ratio (252-254).
Vasodilation of the coronary arteries, especially at or adjacent
to sites of recent plaque disruption, may be particularly beneficial
for the patient with acute infarction. Nitrate-induced vasodilatation
may also have particular utility in those rare patients with coronary
spasm presenting as STEMI.
Clinical trial results have suggested only a modest benefit from
nitroglycerin used acutely in STEMI and continued subsequently.
A pooled analysis of more than 80 000 patients treated with nitrate-like
preparations intravenously or orally in 22 trials revealed a mortality
rate of 7.7% in the control group, which was reduced to 7.4% in
the nitrate group. These data are consistent with a possible small
treatment effect of nitrates on mortality such that 3 to 4 fewer
deaths would occur for every 1000 patients treated (152).
Nitroglycerin may be administered to relieve ischemic pain and is
clearly indicated as a vasodilator in patients with STEMI associated
with LV failure. Nitrates in all forms should be avoided in patients
with initial systolic blood pressures less than 90 mm Hg or greater
than or equal to 30 mm Hg below baseline, marked bradycardia or
tachycardia (256), or known or
suspected RV infarction. Patients with RV infarction are especially
dependent on adequate RV preload to maintain cardiac output and
may experience profound hypotension during administration of nitrates
(257). Phosphodiesterase inhibitors
potentiate the hypotensive effects of nitrates because of their
mechanism of action in releasing nitric oxide and increasing cyclic
guanosine monophosphate (258).
Therefore, it is useful clinical practice to ascertain whether such
agents have been used, and nitrates should not be administered to
patients who have received a phosphodiesterase inhibitor for erectile
dysfunction in the prior 24 hours (48 hours for tadalafil).
Nitroglycerin is commonly given sublingually at doses of 0.4 mg
when patients present with STEMI. Arterial pressure may decline
precipitously because of limited control of the initial dose and
rate of absorption. An intravenous infusion of nitroglycerin allows
clinicians to titrate the therapy in response to the patient’s
blood pressure. A useful intravenous nitroglycerin regimen employs
an initial infusion rate of 5 to 10 mcg per minute with increases
of 5 to 20 mcg per minute until symptoms are relieved or mean arterial
blood pressure is reduced by 10% of its baseline level in normotensive
patients and by up to 30% for hypertensive patients, but in no case
below a systolic pressure of 90 mm Hg or a drop greater than 30
mm Hg below baseline. In view of their marginal treatment benefits,
nitrates should not be used if hypotension limits the administration
of beta-blockers, which have more powerful salutary effects.
6.3.1.3. Analgesia
Class I
Morphine sulfate (2 to 4 mg IV with increments of 2 to 8 mg IV repeated
at 5- to 15-minute intervals) is the analgesic of choice for management
of pain associated with STEMI. (Level of Evidence: C)
Pain relief is an important element in the early management of the
patient with STEMI. There is a tendency to underdose patients with
STEMI because of the desire to assess the response to anti-ischemic
or reperfusion therapy. This should be avoided, because patients
with STEMI have a hyperadrenergic state particularly early after
the onset of coronary occlusion. Conversely, it should not be assumed
that resolution of discomfort after administration of analgesics
indicates reperfusion has occurred (see Section
6.3.1.6.3.7 for further discussion). Pain, which is commonly
severe in the acute phase of the event, contributes to increased
sympathetic activity.
Pain management should be directed toward acute relief of symptoms
of ongoing myocardial ischemia and necrosis and toward general relief
of anxiety and apprehension, the latter of which can heighten pain
perception. Surges of catecholamines have been implicated as having
a role in plaque fissuring and thrombus propagation and in reducing
the threshold for ventricular fibrillation (259).
Because the pain of STEMI is related to ongoing ischemia, interventions
that affect the oxygen supply-demand relationship (i.e., by either
increasing
supply or decreasing demand) may lessen the pain of STEMI (260).
Control of cardiac pain is typically accomplished with a combination
of nitrates, opiate analgesic agents, oxygen, and beta-adrenergic
blockers. Treatment with these agents extends from the ED to the
critical care unit. An important consideration when using intravenous
nitrates is not to lower blood pressure to a level that would preclude
adequate dosage of morphine sulfate for pain control. Morphine sulfate
remains the analgesic agent of choice for management of pain associated
with STEMI, except in documented cases of morphine sensitivity.
The dose required for adequate pain relief varies in relation to
age and body size, as well as blood pressure and heart rate. Anxiety
reduction secondary to morphine administration reduces the patient’s
restlessness and the activity of the autonomic nervous system, with
a consequent reduction of the heart’s metabolic demands. Morphine
administration for patients with pulmonary edema is clearly beneficial
and may promote peripheral arterial and venous dilation, reducing
the work of breathing and slowing the heart rate secondary to combined
withdrawal of sympathetic tone and augmentation of vagal tone (259,260).
Side effects of morphine administration such as hypotension can
be minimized by keeping the patient supine and elevating the lower
extremities if systolic pressure goes below 100 mmHg systolic, assuming
pulmonary edema is not present. The concomitant use of atropine
in 0.5- to 1.5-mg doses intravenously may be helpful in reducing
the excessive vagomimetic effects of morphine if significant bradycardia
or hypotension occurs. Although respiratory depression is relatively
uncommon, patients’ respirations should be monitored, particularly
as their cardiovascular status improves. The narcotic reversing
agent naloxone, 0.1 to 0.2 mg intravenously, can be given initially
if indicated and repeated after 15 minutes if necessary. Nausea
and vomiting as potential side effects of large doses of morphine
may be treated with a phenothiazine (260).
See “Hospital Management”
(Section 7.2.4) for additional discussion of analgesia.
6.3.1.4. Aspirin
Class I
Aspirin should be chewed by patients who have not taken aspirin
before presentation with STEMI. The initial dose should be: 162
mg (Level of Evidence: A) to 325 mg (Level of Evidence:
C). Although some trials have used enteric-coated aspirin for
initial dosing, more rapid buccal absorption occurs with non–enteric-coated
aspirin formulations.
At a dose of 162 mg or more, aspirin produces a rapid clinical
antithrombotic effect caused by immediate and near-total inhibition
of thromboxane A2 production. The Second International Study of
Infarct Survival (ISIS-2) has shown conclusively the efficacy of
aspirin alone for treatment of evolving acute MI, with an absolute
risk difference in 35-day mortality of 2.4% (relative risk reduction
[RRR] 23%) (261). When aspirin
was combined with streptokinase, the absolute risk difference in
mortality was 5.2% (RRR 42%). A metaanalysis demonstrated that aspirin
reduced coronary reocclusion and recurrent ischemic events after
fibrinolytic therapy with either streptokinase or alteplase (262).
Accordingly, aspirin now forms part of the early management of all
patients with suspected STEMI and should be given promptly, certainly
within the first 24 hours, at a dose between 162 and 325 mg and
continued indefinitely at a daily dose of 75 to 162 mg (263).
Although some trials have used entericcoated aspirin for initial
dosing, more rapid buccal absorption occurs with non–enteric-coated
formulations (264).
Unlike fibrinolytic agents, there is little evidence for a time-dependent
effect of aspirin on early mortality. However, data do support the
contention that a chewable aspirin is absorbed more quickly than
one swallowed in the early hours after infarction, particularly
after opiate therapy. The use of aspirin is contraindicated in those
with a hypersensitivity to salicylate. Aspirin suppositories (300
mg) can be used safely and are the recommended route of administration
for patients with severe nausea and vomiting or known upper-gastrointestinal
disorders. In patients with true aspirin allergy (hives, nasal polyps,
bronchospasm, or anaphylaxis), clopidogrel or ticlopidine may be
substituted.
6.3.1.5. Beta-Blockers
Class I
Oral beta-blocker therapy should be administered promptly to those
patients without a contraindication, irrespective of concomitant
fibrinolytic therapy or performance of primary PCI. (Level of
Evidence: A)
Class IIa
It is reasonable to administer IV beta-blockers promptly to STEMI
patients without contra-indications, especially if a tachyarrhythmia
or hypertension is present. (Level of Evidence: B)
During the first few hours after the onset of STEMI, betablocking
agents may diminish myocardial oxygen demand by reducing heart rate,
systemic arterial pressure, and myocardial contractility. In addition,
prolongation of diastole caused by a reduction in heart rate may
augment perfusion to ischemic myocardium, particularly the subendocardium.
As a result, immediate beta-blocker therapy appears to reduce 1)
the magnitude of infarction and incidence of associated complications
in subjects not receiving concomitant fibrinolytic therapy, 2) the
rate of reinfarction in patients receiving fibrinolytic therapy,
and 3) the frequency of life-threatening ventricular tachyarrhythmias.
In patients not receiving fibrinolytic therapy, intravenously administered
beta-blocking agents exert a modestly favorable influence on infarct
size (265). Large early trials
suggested a mortality benefit as well. In ISIS-1 (266),
more than 16 000 patients with suspected acute MI were enrolled
within 12 hours of onset of symptoms; immediate atenolol, 5 to 10
mg IV, followed by oral atenolol, 100 mg daily, reduced 7-day mortality
from 4.3% to 3.7% (p less than 0.02; 6 lives saved per 1000 treated).
The mortality difference between those receiving and not receiving
atenolol was evident by the end of day 1 and was sustained subsequently.
In the Metoprolol In Acute Myocardial Infarction (MIAMI) trial (267),
more than 5700 subjects with evolving MI were randomly assigned
to receive placebo or metoprolol, up to 15 mg IV in 3 divided doses
followed by 50 mg orally every 6 hours for 48 hours and then 100
mg twice per day thereafter. Fifteen-day mortality was reduced with
metoprolol from 4.9% to 4.3%. As in ISIS-1, the mortality difference
between those given placebo and those receiving metoprolol was evident
by the end of day 1, after which it was sustained.
In subjects receiving concomitant fibrinolytic therapy, intravenously
administered beta-blocking drugs diminish the incidence of subsequent
nonfatal reinfarction and recurrent ischemia. In addition, they
may reduce mortality if given particularly early (i.e., within 2
hours) after onset of symptoms. In the Thrombolysis In Myocardial
Infarction Phase II (TIMI-II) trial (268),
in which all patients received IV alteplase, those randomly assigned
to receive metoprolol, 15 mg IV, followed by oral metoprolol, 50
mg twice per day for 1 day and then 100 mg twice per day thereafter,
had a diminished incidence of subsequent nonfatal reinfarction and
recurrent ischemia compared with those begun on oral metoprolol
6 days after the acute event. Among those treated especially early
(i.e., within 2 hours of symptom onset), the composite end point,
death or reinfarction, occurred less often in those given immediate
IV metoprolol than in those who did not receive it.
The benefits of routine early IV use of beta-blockers in the fibrinolytic
era have been challenged by 2 later randomized trials of IV beta-blockade
(269,270)
and by a post hoc analysis of the use of atenolol in the GUSTO-I
trial (271). A subsequent systematic
review of early beta-blocker therapy in STEMI found no significant
reduction in mortality (67). Therefore,
data on the early use of intravenous beta-blockade in STEMI are
inconclusive, and patterns of use vary.
Beta-blockers should not be administered to patients with STEMI
precipitated by cocaine use because of the risk of exacerbating
coronary spasm (272). If IV beta-blockade
induces an untoward effect, such as atrioventricular (AV) block,
excessive bradycardia, or hypotension, the condition is quickly
reversed by infusion of a beta-adrenergic agonist (i.e., isoproterenol
1 to 5 mcg/min). The presence of moderate LV failure early in the
course of STEMI should preclude the use of early IV beta-blockade
until the heart failure has been compensated but is a strong indication
for the oral use of beta-blockade before discharge from the hospital.
The following are relative contraindications to beta-blocker therapy:
heart rate less than 60 bpm, systolic arterial pressure less than
100 mm Hg, moderate or severe LV failure, signs of peripheral hypoperfusion,
shock, PR interval greater than 0.24 second, second- or third-degree
AV block, active asthma, or reactive airway disease.
Randomized trials of beta-blocker therapy in patients with STEMI
undergoing PCI without fibrinolytic therapy have not been performed.
However, it seems reasonable pending further information to extrapolate
data from those receiving another form of revascularization, fibrinolytic
therapy, to the PCI population. The more contemporary CAPRICORN
(Carvedilol Post-infarct Survival Controlled Evaluation) trial (273),
which includes patients undergoing either form of revascularization,
confirms the benefits of beta-blocker therapy in patients with transient
or sustained postinfarction LV dysfunction.
6.3.1.6. Reperfusion
6.3.1.6.1. General Concepts
Class I
All STEMI patients should undergo rapid evaluation for reperfusion
therapy and have a reperfusion strategy implemented promptly after
contact with the medical system. (Level of Evidence: A)
Although rapid spontaneous reperfusion of the infarct artery may
occur, in the majority of patients there is persistent occlusion
of the infarct artery in the first 6 to 12 hours while the affected
myocardial zone is undergoing necrosis. Prompt and complete restoration
of flow in the infarct artery can be achieved by pharmacological
means (fibrinolysis), PCI (balloon angioplasty with or without deployment
of an intracoronary stent under the support of pharmacological measures
to prevent thrombosis), or surgical measures (Figure
3) (24-40).
Despite the extensive improvement in intraoperative preservation
with cardioplegia and hypothermia and in numerous surgical techniques,
it is not logistically possible to provide surgical reperfusion
in a timely fashion, and therefore patients with STEMI who are candidates
for reperfusion routinely receive either fibrinolysis or a catheter-based
treatment.
Evidence exists that expeditious restoration of flow in the obstructed
infarct artery after the onset of symptoms in patients with STEMI
is a key determinant of short- and longterm outcomes regardless
of whether reperfusion is accomplished by fibrinolysis or PCI (24,274,275).
As discussed previously (Section
4.1), efforts should be made to shorten the time from recognition
of symptoms by the patient to contact with the medical system. All
healthcare providers caring for patients with STEMI from the point
of entry into the medical system must recognize the need for rapid
triage and implementation of care in a fashion analogous to the
handling of trauma patients. When considering recommendations for
timely reperfusion of patients with STEMI, the Writing Committee
reviewed data from clinical trials, focusing particular attention
on enrollment criteria for selection of patients for randomization,
actual times reported in the trial report rather than simply the
allowable window specified in the trial protocol, treatment effect
of the reperfusion strategy on individual components of a composite
primary end point (e.g., mortality, recurrent nonfatal infarction),
ancillary therapies (e.g., antithrombin and antiplatelet agents),
and the interface between fibrinolysis and referral for angiography
and revascularization. When available, data from registries were
also reviewed to assess the generalizability of observations from
clinical trials of reperfusion to routine practice. Despite the
wealth of reports on reperfusion for STEMI, it is not possible to
produce a simple algorithm given the heterogeneity of patient profiles
and availability of resources in various clinical settings at various
times of day. This section introduces the recommendations for an
aggressive attempt to minimize the time from entry into the medical
system to implementation of a reperfusion strategy using the concept
of medical system goals. More detailed discussion of these goals
and the issues to be considered in selecting the type of reperfusion
therapy may be found in Section 6.3.1.6.2,
followed by a discussion of available resources in Section
6.3.1.6.2.1.
The medical system goal is to facilitate rapid recognition and treatment
of patients with STEMI such that door-to-needle (or medical contact–to-needle)
time for initiation of fibrinolytic therapy can be achieved within
30 minutes or that door-to-balloon (or medical contact–to-balloon)
time for PCI can be kept under 90 minutes. These goals may not be
relevant for patients with an appropriate reason for delay, such
as uncertainty about the diagnosis (particularly for the use of
fibrinolytic therapy), need for evaluation and treatment of other
life-threatening conditions (e.g., respiratory failure), or delays
associated with the patient’s informed choice to have more
time to consider the decision. In the absence of such types of circumstances,
the emphasis is on having a system in place such that when a patient
with STEMI presents for medical care, reperfusion therapy can be
provided as soon as possible within these time periods. Because
there is not considered to be a threshold effect for the benefit
of shorter timesto reperfusion, these goals should not be understood
as “ideal” times but rather the longest times that should
be considered acceptable. Systems that are able to achieve even
more rapid times for patients should be encouraged. Also, this goal
should not be perceived as an average performance standard but as
a goal that an early treatment system in every hospital should seek
for every appropriate patient.
A critically important goal of reperfusion is to restore flow in
the infarct artery as quickly and as completely as possible, but
the ultimate goal of reperfusion in STEMI is to improve myocardial
perfusion in the infarct zone. Despite adequate restoration of flow
in the epicardial infarct artery, perfusion of the infarct zone
may still be compromised by a combination of microvascular damage
and reperfusion injury (276-278).
Microvascular damage occurs as a consequence of downstream embolization
of platelet microemboli and thrombi followed by the release of substances
from activated platelets that promote occlusion or spasm in the
microvasculature. Reperfusion injury results in cellular edema,
free radical formation, calcium overload, and acceleration of the
apoptotic process. Cytokine activation in the infarct zone leads
to neutrophil accumulation and inflammatory mediators that contribute
to tissue injury.
Thus, construction of an ideal reperfusion regimen in patients with
STEMI not only should focus on the primary means of restoring flow
in the epicardial infarct artery (pharmacological or catheter-based)
but should also include adjunctive and ancillary treatments that
minimize the amount of microvascular damage and protect the jeopardized
myocardial infarct zone that contains cells in various stages of
ischemia, necrosis, and apoptosis (279,280).
The Writing Committee endorses further research to identify the
optimum strategies for achieving these goals.
6.3.1.6.2. Selection Of Reperfusion
Strategy
The literature provides very strong evidence that among patients
with suspected STEMI and without contraindications, the prompt use
of reperfusion therapy is associated with improved survival (156).
Despite such strong evidence, studies continue to indicate that
reperfusion therapy is underutilized and often not administered
soon after presentation (281-283).
Indecision about the choice of reperfusion therapy should not deter
physicians from using these strategies or delay them in administering
therapy.
There is controversy about which form of reperfusion therapy is
superior in various clinical settings. Part of the uncertainty derives
from the continual introduction of new agents, devices, and strategies,
which quickly make previous studies less relevant to contemporary
practice. With pharmacological reperfusion therapies, there are
new agents, dosing regimens, adjunctive treatments, and combined
strategies with procedures that are in a continual process of refinement
and evaluation. Similarly, with catheter-based approaches, there
are new devices, adjunctive therapies, technologies, and combined
strategies with medications that are being introduced and evaluated.
As a result, the evidence base regarding the best approach to reperfusion
therapy is quite dynamic.
Several issues should be considered in selecting the type of reperfusion
therapy, as discussed below.
Time From Onset of Symptoms. Time from onset of symptoms to
fibrinolytic therapy is an important predictor of MI size and patient
outcome (284). The efficacy of
fibrinolytic agents in lysing thrombus diminishes with the passage
of time (279). Fibrinolytic therapy
administered within the first 2 hours (especially the first hour)
can occasionally abort MI and dramatically reduces mortality (Figure
13) (156,159).
The National Heart Attack Alert Working Group (179)
recommends that EDs strive to achieve a 30-minute door-toneedle
time to minimize treatment delays. Prehospital fibrinolysis reduces
treatment delays by up to 1 hour and reduces mortality by 17% (285).
The amount of myocardium at risk, presence of collateral blood flow,
and duration of coronary occlusion are major determinants of myocardial
infarct size (286-289).
In animal models (18), occlusions
persisting greater than 30 minutes produce myonecrosis. Reperfusion
at 90 minutes salvages approximately half of the myocardium at risk.
Myocardial salvage is minimal after 4 to 6 hours of ischemia unless
ischemic preconditioning and/or collateral flow have modified the
wave front of necrosis.
A time-dependent decrease in efficacy of fibrinolytic therapy may
also contribute to the higher mortality rate in patients with longer
symptom duration (279). In contrast,
the ability to produce a patent infarct artery is much less dependent
on symptom duration in patients undergoing PCI. Several reports
claim no influence of time delay on mortality rates when PCI is
performed after 2 to 3 hours of symptom duration (290,291).
One study suggests that time to PCI is only important for patients
presenting with shock (292). Another
showed that time was associated with outcome in higher-risk but
not lower-risk patients (293).
Conversely, others have reported increasing mortality rates with
increasing door-to-balloon times (294,295).
Importantly, after adjustment for baseline characteristics, time
from symptom onset to balloon inflation is significantly correlated
with 1- year mortality in patients undergoing primary PCI for STEMI
(relative risk [RR] equals 1.08 for each 30 minute delay from symptom
onset to balloon inflation, p equals 0.04) (275,275a).
Interestingly, although the CAPTIM (173)
and PRAGUE-2 (177) studies reached
different conclusions about the overall superiority of PCI over
fibrinolysis, important observations were made in the subset of
patients presenting very early after the onset of symptoms. In the
subset of patients presenting within 3 hours of the onset of symptoms
in PRAGUE-2, mortality was equivalent in those treated with streptokinase
and those transferred with PCI (177). Patients
treated within 2 hours of symptom onset in CAPTIM had improved outcomes
with prehospital tissue plasminogen activator (tPA) versus transfer
for PCI (176). (See Section
6.3.1.6.2.1.)
It is also possible that time-to-treatment analyses have been confounded
by other variables (293, 296).
First, higherrisk patients report later to the hospital and may
respond better to PCI than to fibrinolytic agents. Second, shorter
doorto- balloon times may be a surrogate for better quality of care
and adherence to treatment guidelines. The Task Force on the Management
of Acute Myocardial Infarction of the European Society of Cardiology
(297) and this Committee both recommend
a target medical contact– or door-to-balloon time of less
than 90 minutes.
Risk of STEMI. Several models have been developed that
assist clinicians in estimating the risk of mortality in patients
with STEMI (240-242,298,299).
Although these models vary somewhat in the factors loaded into the
risk prediction tool and also vary with respect to statistical measures
of their discriminative power (e.g., C statistic), all the models
provide clinicians with a means to assess the continuum of risk
from STEMI. None of the models have been tested prospectively by
randomizing patients to a reperfusion strategy based on estimated
mortality at presentation. Retrospective analyses do suggest that
the absolute difference in mortality at 30 days between PCI and
fibrinolysis increases in favor of PCI as the estimated risk of
mortality with fibrinolysis increases (300).
Conversely, as the estimated mortality benefit with fibrinolysis
decreases, the absolute mortality benefit of PCI decreases, with
equipoise appearing (i.e., similar 30-day mortality rates) when
the estimated mortality with fibrinolysis is approximately 2% to
3% (300).
When the estimated mortality with fibrinolysis is extremely high,
as is the case in patients with cardiogenic shock, compelling evidence
exists that favors a PCI strategy. The SHOCK trial (SHould we emergently
revascularize Occluded Coronaries for cardiogenic shocK?) demonstrated
that patients with cardiogenic shock have a better 1-year survival
if they have undergone early coronary revascularization (184).
At 1 year, patients in the early revascularization group had a mortality
rate of 53% compared with 66% for the group that had initial medical
stabilization followed by no or late revascularization (184,301).
Observational data from NRMI suggest superiority of PCI over fibrinolysis
for patients with Killip class greater than or equal to II (302).
Risk
of Bleeding. Choice of reperfusion therapy is also affected
by the patient’s risk of bleeding. When both types of reperfusion
are available, the higher the patient’s risk of bleeding with
fibrinolytic therapy, the more strongly the decision should favor
PCI. If PCI is unavailable, then the
benefit of pharmacological reperfusion therapy should be balanced
against the risk. A decision analysis suggested that fibrinolytic
therapy should be favored against no reperfusion treatment until
the risk of a life-threatening bleed exceeds 4% in older patients
who have a risk profile similar to those in the classic randomized
trials of fibrinolytic therapy (247).
Risk
scores for bleeding after fibrinolytic therapy allow for the calculation
of this risk (246). Because they are
derived from less restricted populations, the scores that are most
generalizable are those derived from observational studies (246).
Time Required for Transport to Skilled PCI Laboratory.
The availability of interventional cardiology facilities is a key
determinant of whether PCI can be provided. For facilities that
can offer PCI, the literature suggests that this approach is superior
to pharmacological reperfusion (303).
The trials comparing pharmacological and PCI strategies, however,
were conducted before the advent of more recent pharmacological
and PCI strategies. When a composite end point of death, nonfatal
recurrent MI, or stroke is analyzed, much of the superiority of
a PCI strategy is driven by a reduction in the rate of nonfatal
recurrent MI (Figure 14) (40).
The rate of nonfatal recurrent MI can be influenced both by the
adjunctive therapy used (Figure 3)
(24-40)
and by the proportion of patients who are referred for PCI when
the initial attempt at fibrinolysis fails or myocardial ischemia
recurs after initially successful pharmacological reperfusion (Figure
14) (155).
The experience and location of the PCI laboratory also plays a role
in the choice of therapy. The trials were performed in centers with
highly experienced teams, and their results may not be generalizable
to all PCI laboratories throughout the country. Not all laboratories
can provide prompt, high-quality primary PCI. Even centers with
interventional cardiology facilities may not be able to provide
the staffing required for 24-hour coverage of the catheterization
laboratory. Despite staffing availability, the volume of cases in
the laboratory may be insufficient for the team to acquire and maintain
skills required for rapid PCI reperfusion strategies. A study from
NRMI investigated the effect of volume on the outcomes of patients
treated with PCI versus pharmacological reperfusion strategies (303).
They studied 446 acute-care hospitals, with 112 classified as low-volume
(fewer than or equal to 16 procedures), 223 as intermediatevolume
(17 to 48 procedures), and 111 as high-volume (49 or more procedures)
based on their annual primary angioplasty volume. They reported
that patients hospitalized at intermediate- and high-volume centers
had lower mortality with PCI reperfusion, whereas in the low-volume
centers, there was no significant difference between the 2 reperfusion
strategies. In another article from the NRMI investigators, the
volume of primary PCI procedures, but not pharmacological treatment,
was inversely associated with the mortality rate for patients with
STEMI (304).
A decision must be made when a STEMI patient presents to a center
without interventional cardiology facilities. Fibrinolytic therapy
can generally be provided sooner than primary PCI (Figure
7) (180). As the time delay
for performing PCI increases, the mortality benefit associated with
expeditiously performed primary PCI over fibrinolysis decreases
(305). Compared with a fibrin-specific
lytic agent, a PCI strategy may not reduce mortality when a delay
greater than 60 minutes is anticipated versus immediate administration
of a lytic (Figure 15) (305).
The
balance of risk/benefit between the transfer of patients for PCI
and more immediate treatment with fibrinolytic therapy remains uncertain.
The DANAMI-2 trial (DANish trial in Acute Myocardial Infarction),
conducted in Denmark, found that patients treated at facilities
without interventional cardiology capabilities had better composite
outcomes with transfer for PCI within 2 hours of presentation than
with pharmacological reperfusion treatment at the local hospital
(306). Whether these results could
be replicated elsewhere is not known. An alternative to transfer
is for hospitals without on-site cardiac surgery to develop the
capability to provide primary mechanical reperfusion therapy. A
study by Aversano and colleagues with 11 hospitals in Massachusetts
and Maryland suggested that this approach may improve outcomes (307).
It can be expected, however, that only a limited number of hospitals
could develop such a program, and it has yet to be determined whether
a certain volume of cases would be necessary to maintain the effectiveness
of the service. The economic implications of expansion of the number
of PCI-capable centers that are able to maintain an inventory of
the necessary catheters and other devices and provide 24-hour coverage,
7 days per week, deserve further evaluation from the perspectives
of individual institutions and the global healthcare delivery system.
See additional discussion in Section 6.3.1.6.2.1.
Given the current literature, it is not possible to say definitively
that a particular reperfusion approach is superior for all patients,
in all clinical settings, at all times of day (173,176,177)
(Danchin N; oral presentation, American Heart Association 2003 Annual
Scientific Sessions, Orlando, FL, November 2003). The main point
is that some type of reperfusion therapy should be selected for
all appropriate patients with suspected STEMI. The appropriate and
timely use of some reperfusion therapy is likely more important
than the choice of therapy, given the current literature and the
expanding array of options. Clinical circumstances in which fibrinolytic
therapy is generally preferred or an invasive strategy is generally
preferred are shown in Table 11.
6.3.1.6.2.1. Available Resources
Class I
STEMI patients presenting to a facility without the capability for
expert, prompt intervention with primary PCI within 90 minutes of
first medical contact should undergo fibrinolysis unless contraindicated.
(Level of Evidence: A)
The preferred reperfusion therapy for STEMI must take into account
the location of the patient, the response time and expertise of
the paramedical/ambulance personnel, their relationship to the regional
healthcare facility(s), and the availability, capability, and expertise
of the medical personnel at the facility. The most recent NRMI data
continue to support advantages of primary PCI versus fibrinolysis
in high-PCI–volume hospitals but not among those institutions
with low volume (fewer than or equal to 16 procedures per year)
(303). Approximately 20% of US
hospitals have cardiac catheterization laboratories; and less than
that number have the capacity for primary PCI. The C-PORT study
(Atlantic Cardiovascular Patient Outcomes Research Team), which
randomized 451 fibrinolysis-eligible patients with STEMI treated
in 11 community hospitals with diagnostic catheterization but not
onsite PCI facilities, is of interest. Patients were randomized
within 12 hours of symptom onset to an accelerated alteplase regimen
(median door-to-lytic time was 46 minutes) versus primary PCI (median
door-to-balloon time 101.5 minutes). At 6 months, the incidence
of death, re-MI, and stroke was 12.4% for PCI and 19.9% for fibrinolytic
therapy (p equals 0.03). Because only 18% of the intended sample
size was actually enrolled, this study is significantly underpowered,
and its conclusion can only be hypothesis-generating rather than
definitive (307). Importantly,
most C-PORT patients were randomized between 0800 and 1600 hours,
an experience consistent with NRMI data. The NRMI report also demonstrated
a substantially longer door-to-balloon time when patients with STEMI
undergo direct PCI outside of daylight hours (308).
The Zwolle group evaluated 1702 consecutive patients and found that
the 47% of patients who presented outside “routine duty hours”
(i.e., 1800 to 0800 hours) had a higher rate of both PCI failure
and 30-day mortality than those within the 0800 to 1800 period (6.9%
and 4.2% versus 3.8% and 1.9%, respectively; p less than 0.01) (309).
The reasons for these differences are unclear but could relate to
both patient and process-of-care factors, including variations in
both cognitive function and manual dexterity in sleep-deprived healthcare
providers (308,310).
Two studies germane to STEMI care and resource utilization are noteworthy.
The first, CAPTIM, a comparison of angioplasty and prehospital fibrinolysis
(accelerated alteplase) in STEMI, fell short of its planned 1200-patient
enrollment, and hence was underpowered (173).
Eight hundred forty patients were randomized to prehospital fibrinolysis
versus primary PCI. The primary end point was the composite of all-cause
mortality, nonfatal recurrent MI, and nonfatal disabling stroke
at 30 days, which occurred in 8.2% of patients assigned fibrinolytic
therapy and 6.2% of patients assigned to PCI (p not significant
[NS]). The components for death, reinfarction, and disabling stroke
were 3.8%, 3.7%, and 1.0% for fibrinolytic therapy and 4.8%, 1.7%,
and 0% for PCI. Unlike C-PORT, this trial liberally used rescue
angioplasty (28%), which probably accounts for the relatively low
reinfarction rate in the lytic-treated group. A subsequent analysis
from CAPTIM of the 55% of patients treated within 2 hours of symptom
onset revealed a mortality trend in favor of prehospital fibrinolysis
versus primary PCI (2.2% versus 5.7%, p equals 0.058), whereas those
patients treated beyond 2 hours had a 5.9% versus 3.7% (p equals
0.47) 30-day mortality rate, respectively (176).
Interestingly, there was a significant reduction in the frequency
of cardiogenic shock for patients treated within 2 hours with prehospital
fibrinolysis (1.3% versus 5.3%, p equals 0.032), whereas the frequency
of this event after 2 hours was similar (i.e., 3.9% versus 4.4%,
respectively) (176).
The
DANAMI-2 study, which compared primary PCI versus accelerated alteplase,
enrolled 1572 patients versus the 2000 patients planned (306).
Patients were eligible if they had a sum of greater than 0.4 mV
of ST elevation in 2 contiguous leads on their presenting ECG within
12 hours of symptom onset; however, patient enrollment consisted
of 37% of those screened, and patients deemed to be at high risk
during ambulance transport were excluded (306).
Twentynine hospitals, of which 5 conducted primary PCI and were
located a mean of 35 miles from referring hospitals (maximum 95
miles), participated. The median door-to-needle time for patients
randomized to fibrinolysis was approximately 50 minutes for patients
presenting either to a community (referral) hospital or an invasive
center. For patients who presented to a community hospital (where
1129 patients were enrolled), the time from initial presentation
to balloon inflation at an invasive center was 108 minutes; the
door-toballoon time was 93 minutes for patients presenting to an
invasive center and randomized to PCI. The primary composite end
point of death, reinfarction, and stroke through 30 days occurred
in 14.2% of the fibrinolysis-treated patients and 8.5% of the PCI-treated
patients (p less than 0.001). The individual end-point components
of death, reinfarction, and stroke occurred in 7.8%, 6.3%, and 2.0%
of the fibrinolysistreated patients and 6.6%, 1.6%, and 1.1% of
the PCI-treated patients, respectively. In addition to the exclusion
of highrisk patients for transport, some caveats in DANAMI-2 are
noteworthy: 1) The antithrombotic dosing regimen was in excess of
ACC/AHA guidelines. 2) The protocol specified that repeat fibrinolysis
was to be used for failed reperfusion, reinfarction, and recurrent
ST-elevation ischemia; this strategy was used in 26 patients within
12 hours after randomization, and only 1.9% underwent rescue PCI.
3) Patients with prior stroke were included, and an imbalance in
this baseline characteristic was present, (i.e., 4.0% for fibrinolysis
versus 2.7% for PCI ([1-sided p equals 0.06]). 4) The difference
in reinfarction rates between the 2 groups was likely exaggerated
by the exclusion of those patients associated with invasive procedures
(311).
Hence, on the basis of the data, patients with STEMI presenting
to a facility without the capability for expert, prompt intervention
with primary PCI within 90 minutes of first medical contact should
undergo fibrinolysis unless contraindicated. (See Sections
6.3.1.6.4.2, Primary PCI, and 6.3.1.6.4.2.4,
Interhospital Transfer for Primary PCI.)
6.3.1.6.3. Pharmacological Reprefusion.
Rationale For Fibrinolytic Therapy. Although the clinical
features of coronary obstruction were described nearly a century
ago (312,313),
thrombotic obstruction of the infarct artery as a cause of STEMI
was not proven until 1980 (20).
The benefits of fibrinolytic therapy are maximal when there is prompt,
adequate restoration of flow in the epicardial infarct artery and
perfusion of the myocardium in the infarct zone. Controlled clinical
trials have demonstrated the potential forfunctional, clinical,
and mortality benefits only if fibrinolytic therapy is given within
12 hours. (See additional discussion on the use of antithrombins
and antiplatelet agents as ancillary therapy in Sections
6.3.1.6.8.1 and |
