Evaluation, Differential Diagnosis, and Approach to the Patient with Possible Acute Coronary Syndrome

Judd E. Hollander

The standard 12-lead electrocardiogram is the single best test to identify patients with acute myocardial infarction and ST-segment Elevation MI (STEMI) immediately upon ED presentation,¹ despite the fact that it has relatively low sensitivity for detection of acute coronary syndromes.

Significance of history and physical findings in possible ACS patients
The 12-lead electrocardiogram (ECG) as central to evaluation
Utility of cardiac markers in decision making and risk stratification
Imaging modalities in ED patients with chest pain
Disposition from the ED—home, hospital, decision unit

Introduction

Chest pain is the most common potential life-threatening complaint for patients presenting to the emergency department (ED), occurring in approximately 8 million patients annually in the United States.²

Chest pain, shortness of breath, and other symptoms of potential acute coronary syndrome (ACS) present an initial diagnostic challenge. Demographics, traditional risk factors, chest pain characteristics, and physical examination can assist disposition decisions but are insufficient by themselves to identify patients safe for ED release.³^,⁴^,⁵^,⁶^,⁷^,⁸^,⁹ Patients with acute myocardial infarction (AMI) can present with a wide variety of symptoms, making diagnosis difficult.¹⁰^,¹¹^,¹² Some patients may have objective evidence of a clear-cut diagnosis; however, the majority do not.¹³

Physicians maintain a low threshold for admitting patients to exclude an ACS. As a result, the percentage of patients admitted to coronary care units who actually suffer cardiac ischemia has been reported to be as low as 15% to 30% historically, and up to $8 billion to $10 billion a year is spent on hospital stays to rule out AMI and ACS.⁷^,¹²^,¹³^,¹⁴^,¹⁵^,¹⁶ Despite this high rate of admission, up to 2% to 5% of patients with AMI are still inappropriately discharged from the ED, with concomitant increased morbidity and mortality.¹⁰^,¹²^,¹⁵^,¹⁶ Missed AMI ranks as the highest single diagnosis in terms of dollars paid and third highest in terms of frequency of claims in malpractice against emergency physicians.¹⁷

History

A carefully conducted history and physical examination comprise the initial assessment of patients presenting with potential ACS. The precision of clinical features for the evaluation of chest pain is quite variable.¹⁸^,¹⁹^,²⁰^,²¹ Hickan et al. found that features associated with a lower probability of AMI, such as pleuritic, positional, and sharp chest pain, showed poor to fair interphysician reliability (kappa values of 0.27–0.44). High-risk features (radiation to left arm, substernal location, and history of AMI) were more reliable (kappa 0.74–0.89).¹⁸ Thus, history is most reliable in “ruling in” high-risk patients but is less reliable when being used to “rule out” ACS. Likelihood ratios for several clinical features are shown in Table 3-1.

Cardiac Risk Factors

In the presence of symptoms, cardiac risk factors are poor predictors of risk for MI or ACS.⁹^,²² Traditional cardiac risk factors, such as hypertension, diabetes mellitus, tobacco use, family history of coronary artery disease (CAD) at an early age, and hypercholesterolemia, are not predictive of the cardiac risk in ED patients with chest pain. Traditional cardiac risk factors were derived from population-based longitudinal cohort studies of asymptomatic patients. In contrast, ED patients with chest pain have already been identified as being at increased risk by the very fact that they have symptoms.
The presence of symptoms outweighs the predictive abilities of cardiac risk factors. A lack of cardiac risk factors does not sufficiently decrease cardiac risk in ED patients such that they can be expeditiously released from the ED unless they are <40 years old and have normal ECGs.²²^,²³^,²⁴^,²⁵

Table 3-1 • Likelihood Ratios for Clinical Features That Increase or Decrease Risk of AMI in Patients Presenting With Chest Pain

Clinical Feature	Likelihood Ratio (95% CI)
Increased Likelihood of AMI
Described as pressure	1.3 (1.2–1.5)
Pain in chest or left arm	2.7^a
Chest pain radiation
To right arm or shoulder	4.7 (1.9–12)
To left arm	2.3 (1.7–3.1)
To both left and right arms	7.1 (3.6–14.2)
To both arms or shoulders	4.1 (2.5–6.5)
Chest pain most important symptom	2.0^a
Chest pain associated with exertion	2.4 (1.5–3.8)
Worse than previous angina or similar to prior AMI	1.8 (1.6–2.0)
History of MI	1.5–3.0^b
Nausea or vomiting	1.9 (1.7–2.3)
Diaphoresis	2.0 (1.9–2.2)
Third heart sound	3.2 (1.6–6.5)
Hypotension (systolic BP <80 mm Hg)	3.1 (1.8–5.2)
Pulmonary crackles	2.1 (1.4–3.1)
Decreased Likelihood of AMI
Pleuritic chest pain	0.2 (0.1–0.3)
Described as sharp	0.3 (0.2–0.5)
Positional chest pain	0.3 (0.2–0.5)
Reproduced by palpation	0.3 (0.2–0.4)
Inframammary location	0.8 (0.7–0.9)
Not associated with exertion	0.8 (0.6–0.9)
^aData not available to calculate confidence intervals. ^bIn heterogeneous studies the likelihood ratios are reported as ranges. Sources: Adapted from Panju AA, Hemmelgarm BR, Guyatt GH, et al. Is this patient having a myocardial infarction? JAMA 1998;280:1256–1263, and Swap CJ, Nagurney JT. Value and limitations of chest pain history in the evaluation of patients with suspected acute coronary syndromes. JAMA 2005;294:2623–2629.

A lack of cardiac risk factors does not sufficiently decrease cardiac risk in ED patients such that they can be expeditiously released from the ED, unless the patient is less than 40 years old and has a normal electrocardiogram.

Physical Examination

Likewise, the use of the physical examination to distinguish patients with ACS from patients with noncardiac chest pain is suboptimal. Patients with ACS may appear deceptively well without any clinical signs of distress or may be uncomfortable, pale, cyanotic, and in respiratory distress. Heavy reliance on individual physical examination findings is not wise. The first and second heart sounds may be diminished due to poor myocardial contractility, even in the absence of myocardial ischemia. An S3 is present in only 15% to 20% of patients with AMI, but it is also common in those with heart
failure. An S4 is common in patients with long-standing hypertension or myocardial dysfunction with or without acute ischemia. The presence of a murmur can be an ominous sign (flail leaflet of the mitral valve or a ventricular septal defect), or it can reflect long-standing valvular heart disease. Signs and symptoms of congestive heart failure have poor interrater reliability (S3 gallop, kappa = 0.14–0.37; rales, 0.12–0.31; neck vein distention, 0.31–0.51; hepatomegaly, 0.00–0.16; and dependent edema, 0.27–0.64).¹⁹ Thus the use of these individual signs and symptoms as strong guides to management can be misleading in the ED setting.

Atypical Presentations

Atypical presentations and silent myocardial ischemia are common. Some 22% to 40% of patients with Q-wave MI are clinically unrecognized.²⁶ Women and the elderly are more likely to have atypical presentations. The prognosis for patients who have atypical symptoms at the time of their infarction is worse than that of patients who had more typical symptoms.

The response to initial therapy is not useful in “ruling in” or “ruling out” an ACS. Relief of symptoms with either nitroglycerin or a “GI cocktail” occurs in pain related and unrelated to myocardial ischemia.²⁷^,²⁸^,²⁹^,³⁰

Differential Diagnosis

The differential diagnosis of chest pain and shortness of breath is extensive, with ACS being the most common potentially life-threatening condition. Other life-threatening conditions include aortic dissection, pulmonary embolism, tension pneumothorax, pericardial tamponade, and mediastinitis.

Approximately 15% to 30% of patients who present to the ED with nontraumatic chest pain have ACS.⁷^,¹²^,¹³^,¹⁴^,¹⁵^,¹⁶^,³¹^,³² The 28-day case mortality rate for an ACS among patients in developed nations is approximately 10% but varies with the severity of disease and the treatment provided.

Aortic Dissection

The incidence of aortic dissection is estimated at 3 per 100,000 patients per year.³³ Aortic dissection most commonly affects patients with systemic hypertension in their seventh decade of life. The probability of aortic dissection increases when pain is of abrupt onset with a sharp, tearing, and/or ripping character; there is a variation in pulse (absence of a proximal extremity or carotid pulse) and/or blood pressure (difference of >20 mm Hg between the right and left arms), and the chest radiography shows mediastinal or aortic widening.³⁴^,³⁵

Pulmonary Embolism

The incidence of pulmonary embolism (PE) is estimated at over 1 in 1,000 patients, but the diagnosis is often missed and the incidence may be higher.³⁶ Risk stratification depends on the pretest probability for PE. Several scoring systems exist to characterize patient risk for PE, including the Canadian (Wells) score, the Charlotte rule, the Geneva score, and others.³⁷^,³⁸^,³⁹^,⁴⁰ Patients with symptoms suggestive of PE and right ventricular dysfunction or hemodynamic instability are at high risk for PE and major comorbidity and death.

Pneumothorax

Pneumothorax can occur spontaneously or following trauma or pulmonary procedures. Patients with primary spontaneous pneumothorax tend to be younger males who are tall and thin. Secondary spontaneous pneumothorax occurs with greatest frequency in patients with chronic obstructive pulmonary disease, cystic fibrosis, or asthma.⁴¹ Tension pneumothorax is diagnosed clinically by the presence of unilateral decreased breath sounds, hypotension, and shifting of the trachea in the opposite direction.

Mediastinitis

Mediastinitis is less frequent and occurs following odontogenic infections, esophageal perforation, and iatrogenic complications of cardiac surgery or upper gastrointestinal and airway procedures. “Hamman’s crunch,” heard over the anterior chest, is strongly suggestive. Mortality for patients with mediastinitis remains high (14%–42%), even when it is treated with operative debridement and antibiotics.⁴²^,⁴³^,⁴⁴^,⁴⁵

Pericardial Tamponade

Pericardial tamponade results in a direct compromise in cardiac filling, producing a picture resembling cardiogenic shock that requires emergent reduction in pericardial pressure by pericardiocentesis. Tamponade may occur with aortic dissection, after thoracic trauma, or as a consequence of acute pericarditis from infection, malignancy, uremia, or other causes.

Other Conditions

Other conditions are more common but less serious. Gastroesophageal reflux disease, esophageal spasm, hiatal hernias, as well as upper abdominal processes can cause chest pain. Gastrointestinal causes account for the symptoms of a sizable number of patients who complain of chest pain and do not have an ACS. Respiratory infections such as pneumonia and bronchitis are frequently accompanied by chest discomfort but can usually be identified on the basis of cough and fever. Chest tightness is a common complaint in patients with acute exacerbations of asthma, which can often be distinguished by wheezing on examination.

Acute decompensated heart failure can result in chest discomfort that may or may not represent an ACS because heart failure is the result of chronic CAD and other
cardiovascular conditions in some patients.⁴⁶^,⁴⁷ Valvular heart disease, such as aortic stenosis, can result in chest pain that often signifies severe disease, while mitral valve prolapse is associated with chest pain that may have few if any long-term consequences. Infectious or inflammatory causes of chest discomfort include pericarditis, myocarditis, and endocarditis.

Chest heaviness or discomfort may be noted with pleural effusions. Pulmonary malignancy can cause chest pain if there is pleural involvement. Musculoskeletal causes of chest pain include muscle strains and costochondritis as well as overuse syndromes. Herpes zoster can usually be identified as the cause of chest pain, but in rare instances the pain may be evident prior to the exanthem.

Electrocardiogram

The standard 12-lead electrocardiogram (ECG) is the single best test to identify patients with STEMI immediately upon ED presentation,⁷ despite the fact that it has relatively low sensitivity for the detection of ACS. The sensitivity of ST-segment deviation for the detection of AMI is 35% to 50%, leaving more than half of all AMI patients unidentified.⁷^,⁴⁸ However, even among patients with ST-segment-elevation MI (STEMI), ECG variables can further risk-stratify the likelihood of 30-day mortality (Table 3-2).⁴⁹ ECG criteria that increase the risk of AMI in ED patients with chest pain are shown in Tables 3-2 and 3-3.²⁰ Patients with normal or nonspecific ECGs have a 1%–5% incidence of AMI and a 4% to 23% incidence of unstable angina.⁷^,⁴⁸^,⁵⁰^,⁵¹ Patients with nondiagnostic ECGs or with ischemia that is not known to be old have a 4% to 7% incidence of AMI and a 21% to 48% incidence of unstable angina. Demonstration of new ischemia increases the risk of AMI to 25% to 73%, and the unstable angina risk to 14% to 43%.⁷ A normal diagnostic 12-lead ECG cannot conclusively exclude ACS.⁷^,⁵² The ECG should be used in conjunction with clinical history and cardiac markers to determine admission location and treatment for patients with ACS.

Table 3-2 • Multivariate Significance of the Electrocardiogram in Patients With STEMI Enrolled in GUSTO-1

Electrocardiographic Feature	Odds Ratio (95% CI)
Sum of ST-segment deviation (19 vs. 8 mm)	1.53 (1.38–1.69)
Sum of ST-segment decrease (–1 vs. –7 mm)	0.77 (0.72–0.83)
Heart rate (84 vs. 60 bpm)	1.49 (1.41–1.59)
Sum ST-segment increase in II, III, and aVF (6 vs. 0 mm)	0.79 (0.71–0.89)
QRS duration 100 vs. 80 msec
Anterior infarct	1.55 (1.43–1.68)
Other location	1.08 (1.03–1.13)
Anterior infarction
QRS duration 100 msec	1.08 (1.03–1.13)
QRS duration 50 msec	0.61 (0.43–0.86)
Inferior infarction
No prior AMI	0.67 (0.50–0.90)
Prior AMI	1.41 (0.98–2.02)
Prior infarction
Inferior infarction	2.47 (2.02–3.00)
Other location	1.17 (0.98–1.41)
Source: Hathaway WR, Peterson ED, Wagner GS, et al. Prognostic significance of the initial electrocardiogram in patients with acute myocardial infarction. JAMA 1998;279:387–391.

Continuous ECG Monitoring

Because of the relatively poor sensitivity of the standard 12-lead ECG to detect patients with ACS, additional electrocardiographic strategies have been proposed. A continuous 12-lead ECG monitors and records new ECGs every
20 seconds. When the ST-segment baseline is altered, it sets off an alarm and prints a copy of the new ECG. This technology is most often employed in chest pain–observation units, where it might be useful for monitoring patients who present with non-AMI ACS for ECG evidence of injury.⁴⁸ Because of costs, concerns regarding labile ST-segment and T-wave changes from hyperventilation or patient movement, and a lack of ED-based prospective studies demonstrating clinically relevant differences in outcome, continuous 12-lead ECGs have not been recommended for routine use.⁴⁸

Table 3-3 • Electrocardiographic Features Predictive of AMI in Patients With Acute Chest Pain

Electrocardiographic Feature	Likelihood Ratio (95% CI)
New ST-segment elevation ≥1 mm	5.7–53.9^a
New Q wave	5.3–24.8^a
Any ST-segment elevation	11.2 (7.1–17.8)
New conduction defect	6.3 (2.5–15.7)
New ST-segment depression	3.0–5.2^a
Any Q wave	3.9 (2.7–5.7)
Any ST-segment depression	3.2 (2.5–4.1)
T wave peaking and/or inversion ≥1 mm	3.1^b
New T-wave inversion	2.4–2.8^a
Any conduction defect	2.7 (1.4–5.4)
^aIn heterogeneous studies the likelihood ratios are reported as ranges. ^bData not available to calculate confidence intervals. Source: Panju AA, Hemmelgarm BR, Guyatt GH, et al. Is this patient having a myocardial infarction? JAMA 1998;280:1256–1263, with permission.

Additional ECG Leads

Other ECGs with 15, 18, and 22 leads have been studied.⁴⁸^,⁵³ In one study, addition of V4R, V8, and V9 increased the sensitivity to detect ST-segment elevation to 59% without a loss of specificity.⁵³ The addition of V4R to V6R and V7 to V9 as posterior leads led to an 8% increase in sensitivity for AMI relative to a standard 12-lead ECG but at the cost of a 7% decrease in specificity.⁵⁴ Twenty-two–lead and body-surface-mapping ECGs have not been sufficiently studied to recommend their use. Right-sided leads (rV4) should be used in the setting of inferior wall infarction to assess possible RV involvement.⁴⁸

ECG Confounders

There are several clinical conditions where ECG interpretation of ACS is difficult, particularly paced rhythms and left-bundle-branch block (LBBB) (Table 3-4). In the setting of a LBBB, the presence of ST-segment elevation ≥1 mm and concordance with the QRS complex or ST-segment depression ≥1 mm in leads V1, V2, or V3 suggest AMI.⁵⁵ ST-segment elevation ≥5 mm that is discordant with the QRS complex increases the likelihood of AMI but has poor specificity.

Uncomplicated right ventricular pacing causes secondary repolarization changes of opposing polarity to that of the predominant QRS complex. Most leads have predominant negative QRS complexes followed by ST-segment elevation and positive T waves. ST-segment elevation ≥5 mm is most indicative of AMI in leads with predominantly negative QRS complexes. Any ST-segment elevation concordant with the QRS complex in a predominantly positive QRS complex is highly specific for AMI. The QRS complex is predominantly negative in leads V1 to V3. ST-segment depression in these leads had 80% specificity for AMI.⁵⁶

Risk-Stratification Algorithms

Although clinical and computer algorithms can successfully risk-stratify patients, they are not able to reliably identify a group of patients at such low risk of an ACS that they could be safely and immediately released from the Emergency Department.

Several risk-stratification algorithms have been studied and validated.

Table 3-4 • Some Clinical Conditions Where the Electrocardiogram Interpretation Can Be Difficult

May have ST Segment Elevation in the Absence of AMI
Early repolarization
Left ventricular hypertrophy
Pericarditis
Myocarditis
Left ventricular aneurysms
Idiopathic hypertrophic subaortic stenosis (IHSS)
Hypothermia
Paced rhythms
Left bundle-branch block
May have ST-segment Depressions in the Absence of Ischemia
Hypokalemia
Digoxin effect
Cor pulmonale and right heart strain
Early repolarization
Left ventricular hypertrophy
Paced rhythms
Left bundle-branch block
May have T-wave Inversions in the Absence of Ischemia
Persistent juvenile pattern
Stokes–Adams syncope or seizures
Posttachycardia T-wave inversion
Postpacemaker T-wave inversion
Intracranial pathology (CNS bleeds)
Mitral valve prolapse
Pericarditis
Primary or secondary myocardial diseases
Pulmonary embolism or cor pulmonale from other causes
Spontaneous pneumothorax
Myocardial contusion
Left ventricular hypertrophy
Paced rhythms
Left bundle-branch block

Source: Hollander JE. Acute coronary syndromes: unstable angina, myocardial ischemia and infarction. In Tintanelli JE, Kelen GD, Stapczynski JS, eds. Emergency Medicine: A Comprehensive Study Guide, 6th ed. New York: McGraw-Hill, 2003:343–352.

Only gold members can continue reading. Log In or Register to continue