APPROACH TO THE PATIENT: Chest Discomfort
Given the broad set of potential causes and the heterogeneous risk of serious complications in patients who present with acute nontraumatic chest discomfort, the priorities of the initial clinical encounter include assessment of (1) the patient’s clinical stability and (2) the probability that the patient has an underlying cause of the discomfort that may be life-threatening. The high-risk conditions of principal concern are acute cardiopulmonary processes, including ACS, acute aortic syndrome, pulmonary embolism, tension pneumothorax, and pericarditis with tamponade. Among non-cardiopulmonary causes of chest pain, esophageal rupture likely holds the greatest urgency for diagnosis. Patients with these conditions may deteriorate rapidly despite initially appearing well. The remaining population with non-cardiopulmonary conditions has a more favorable prognosis during completion of the diagnostic work-up. A rapid targeted assessment for a serious cardiopulmonary cause is of particular relevance for patients with acute ongoing pain who have presented for emergency evaluation. Among patients presenting in the outpatient setting with chronic pain or pain that has resolved, a general diagnostic assessment is reasonably undertaken (see “Outpatient Evaluation of Chest Discomfort,” below). A series of questions that can be used to structure the clinical evaluation of patients with chest discomfort is shown in Table 19-2. HISTORY
The evaluation of nontraumatic chest discomfort relies heavily on the clinical history and physical examination to direct subsequent diagnostic testing. The evaluating clinician should assess the quality, location (including radiation), and pattern (including onset and duration) of the pain as well as any provoking or alleviating factors. The presence of associated symptoms may also be useful in establishing a diagnosis. Quality of Pain
The quality of chest discomfort alone is never sufficient to establish a diagnosis. However, the characteristics of the pain are pivotal in formulating an initial clinical impression and assessing the likelihood of a serious cardiopulmonary process (Table 19-1), including acs in particular (Fig. 19-2). Pressure or tightness is consistent with a typical presentation of myocardial ischemic pain. Nevertheless, the clinician must remember that some patients with ischemic chest symptoms deny any “pain” but rather complain of dyspnea or a vague sense of anxiety. The severity of the discomfort has poor diagnostic accuracy. It is often helpful to ask about the similarity of the discomfort to previous definite ischemic symptoms. It is unusual for angina to be sharp, as in knifelike, stabbing, or pleuritic; however, patients sometimes use the word “sharp” to convey the intensity of discomfort rather than the quality. Pleuritic discomfort is suggestive of a process involving the pleura, including pericarditis, pulmonary embolism, or pulmonary parenchymal processes. Less frequently, the pain of pericarditis or massive pulmonary embolism is a steady severe pressure or aching that can be difficult to discriminate from myocardial ischemia. “Tearing” or “ripping” pain is often described by patients with acute aortic dissection. However, acute aortic emergencies also present commonly with severe, knifelike pain. A burning quality can suggest acid reflux or peptic ulcer disease but may also occur with myocardial ischemia. Esophageal pain, particularly with spasm, can be a severe squeezing discomfort identical to angina. Location of Discomfort
A substernal location with radiation to the neck, jaw, shoulder, or arms is typical of myocardial ischemic discomfort. Some patients present with aching in sites of radiated pain as their only symptoms of ischemia. However, pain that is highly localized—e.g., that which can be demarcated by the tip of one finger—is highly unusual for angina. A retrosternal location should prompt consideration of esophageal pain; however, other gastrointestinal conditions usually present with pain that is most intense in the abdomen or epigastrium, with possible radiation into the chest. Angina may also occur in an epigastric location. However, pain that occurs solely above the mandible or below the epigastrium is rarely angina. Severe pain radiating to the back, particularly between the shoulder blades, should prompt consideration of an acute aortic syndrome. Radiation to the trapezius ridge is characteristic of pericardial pain and does not usually occur with angina. Pattern
Myocardial ischemic discomfort usually builds over minutes and is exacerbated by activity and mitigated by rest. In contrast, pain that reaches its peak intensity immediately is more suggestive of aortic dissection, pulmonary embolism, or spontaneous pneumothorax. Pain that is fleeting (lasting only a few seconds) is rarely ischemic in origin. Similarly, pain that is constant in intensity for a prolonged period (many hours to days) is unlikely to represent myocardial ischemia if it occurs in the absence of other clinical consequences, such as abnormalities of the ECG, elevation of cardiac biomarkers, or clinical sequelae (e.g., heart failure or hypotension). Both myocardial ischemia and acid reflux may have their onset in the morning, the latter because of the absence of food to absorb gastric acid. Provoking and Alleviating Factors
Patients with myocardial ischemic pain usually prefer to rest, sit, or stop walking. However, clinicians should be aware of the phenomenon of “warm-up angina” in which some patients experience relief of angina as they continue at the same or even a greater level of exertion without symptoms (Chap. 293). Alterations in the intensity of pain with changes in position or movement of the upper extremities and neck are less likely with myocardial ischemia and suggest a musculoskeletal etiology. The pain of pericarditis, however, often is worse in the supine position and relieved by sitting upright and leaning forward. Gastroesophageal reflux may be exacerbated by alcohol, some foods, or by a reclined position. Relief can occur with sitting.
Exacerbation by eating suggests a gastrointestinal etiology such as peptic ulcer disease, cholecystitis, or pancreatitis. Peptic ulcer disease tends to become symptomatic 60–90 min after meals. However, in the setting of severe coronary atherosclerosis, redistribution of blood flow to the splanchnic vasculature after eating can trigger postprandial angina. The discomfort of acid reflux and peptic ulcer disease is usually diminished promptly by acid-reducing therapies. In contrast with its impact in some patients with angina, physical exertion is very unlikely to alter symptoms from gastrointestinal causes of chest pain. Relief of chest discomfort within minutes after administration of nitroglycerin is suggestive of but not sufficiently sensitive or specific for a definitive diagnosis of myocardial ischemia. Esophageal spasm may also be relieved promptly with nitroglycerin. A delay of >10 min before relief is obtained after nitroglycerin suggests that the symptoms either are not caused by ischemia or are caused by severe ischemia, such as during acute MI. Associated Symptoms
Symptoms that accompany myocardial ischemia may include diaphoresis, dyspnea, nausea, fatigue, faintness, and eructations. In addition, these symptoms may exist in isolation as anginal equivalents (i.e., symptoms of myocardial ischemia other than typical angina), particularly in women and the elderly. Dyspnea may occur with multiple conditions considered in the differential diagnosis of chest pain and thus is not discriminative, but the presence of dyspnea is important because it suggests a cardiopulmonary etiology. Sudden onset of significant respiratory distress should lead to consideration of pulmonary embolism and spontaneous pneumothorax. Hemoptysis may occur with pulmonary embolism, or as blood-tinged frothy sputum in severe heart failure but usually points toward a pulmonary parenchymal etiology of chest symptoms. Presentation with syncope or pre-syncope should prompt consideration of hemodynamically significant pulmonary embolism or aortic dissection as well as ischemic arrhythmias. Although nausea and vomiting suggest a gastrointestinal disorder, these symptoms may occur in the setting of MI (more commonly inferior MI), presumably because of activation of the vagal reflex or stimulation of left ventricular receptors as part of the Bezold-Jarisch reflex. Past Medical History
The past medical history is useful in assessing the patient for risk factors for coronary atherosclerosis (Chap. 291e) and venous thromboembolism (Chap. 300) as well as for conditions that may predispose the patient to specific disorders. For example, a history of connective tissue diseases such as marfan syndrome should heighten the clinician’s suspicion of an acute aortic syndrome or spontaneous pneumothorax. A careful history may elicit clues about depression or prior panic attacks. PHYSICAL EXAMINATION
In addition to providing an initial assessment of the patient’s clinical stability, the physical examination of patients with chest discomfort can provide direct evidence of specific etiologies of chest pain (e.g., unilateral absence of lung sounds) and can identify potential precipitants of acute cardiopulmonary causes of chest pain (e.g., uncontrolled hypertension), relevant comorbid conditions (e.g., obstructive pulmonary disease), and complications of the presenting syndrome (e.g., heart failure). However, because the findings on physical examination may be normal in patients with unstable ischemic heart disease, an unremarkable physical exam is not definitively reassuring. General
The patient’s general appearance is helpful in establishing an initial impression of the severity of illness. Patients with acute MI or other acute cardiopulmonary disorders often appear anxious, uncomfortable, pale, cyanotic, or diaphoretic. Patients who are massaging or clutching their chests may describe their pain with a clenched fist held against the sternum (Levine’s sign). Occasionally, body habitus is helpful—e.g., in patients with Marfan syndrome or the prototypical young, tall, thin man with spontaneous pneumothorax. Vital Signs
Significant tachycardia and hypotension are indicative of important hemodynamic consequences of the underlying cause of chest discomfort and should prompt a rapid survey for the most severe conditions, such as acute MI with cardiogenic shock, massive pulmonary embolism, pericarditis with tamponade, or tension pneumothorax. Acute aortic emergencies usually present with severe hypertension but may be associated with profound hypotension when there is coronary arterial compromise or dissection into the pericardium. Sinus tachycardia is an important manifestation of submassive pulmonary embolism. Tachypnea and hypoxemia point toward a pulmonary cause. The presence of low-grade fever is nonspecific because it may occur with MI and with thromboembolism in addition to infection. Pulmonary
Examination of the lungs may localize a primary pulmonary cause of chest discomfort, as in cases of pneumonia, asthma, or pneumothorax. Left ventricular dysfunction from severe ischemia/infarction as well as acute valvular complications of MI or aortic dissection can lead to pulmonary edema, which is an indicator of high risk. Cardiac
The jugular venous pulse is often normal in patients with acute myocardial ischemia but may reveal characteristic patterns with pericardial tamponade or acute right ventricular dysfunction (Chaps. 267 and 288). Cardiac auscultation may reveal a third or, more commonly, a fourth heart sound, reflecting myocardial systolic or diastolic dysfunction. Murmurs of mitral regurgitation or a harsh murmur of a ventricular-septal defect may indicate mechanical complications of STEMI. A murmur of aortic insufficiency may be a complication of proximal aortic dissection. Other murmurs may reveal underlying cardiac disorders contributory to ischemia (e.g., aortic stenosis or hypertrophic cardiomyopathy). Pericardial friction rubs reflect pericardial inflammation. Abdominal
Localizing tenderness on the abdominal exam is useful in identifying a gastrointestinal cause of the presenting syndrome. Abdominal findings are infrequent with purely acute cardiopulmonary problems, except in the case of underlying chronic cardiopulmonary disease or severe right ventricular dysfunction leading to hepatic congestion. Vascular
Pulse deficits may reflect underlying chronic atherosclerosis, which increases the likelihood of coronary artery disease. However, evidence of acute limb ischemia with loss of the pulse and pallor, particularly in the upper extremities, can indicate catastrophic consequences of aortic dissection. Unilateral lower-extremity swelling should raise suspicion about venous thromboembolism. Musculoskeletal
Pain arising from the costochondral and chondrosternal articulations may be associated with localized swelling, redness, or marked localized tenderness. Pain on palpation of these joints is usually well localized and is a useful clinical sign, though deep palpation may elicit pain in the absence of costochondritis. Although palpation of the chest wall often elicits pain in patients with various musculoskeletal conditions, it should be appreciated that chest wall tenderness does not exclude myocardial ischemia. Sensory deficits in the upper extremities may be indicative of cervical disk disease. ELECTROCARDIOGRAPHY
Electrocardiography is crucial in the evaluation of nontraumatic chest discomfort. The ECG is pivotal for identifying patients with ongoing ischemia as the principal reason for their presentation as well as secondary cardiac complications of other disorders. Professional society guidelines recommend that an ECG be obtained within 10 min of presentation, with the primary goal of identifying patients with ST-segment elevation diagnostic of MI who are candidates for immediate interventions to restore flow in the occluded coronary artery. ST-segment depression and symmetric T-wave inversions at least 0.2 mV in depth are useful for detecting myocardial ischemia in the absence of STEMI and are also indicative of higher risk of death or recurrent ischemia. Serial performance of ECGs (every 30–60 min) is recommended in the ED evaluation of suspected ACS. In addition, an ECG with right-sided lead placement should be considered in patients with clinically suspected ischemia and a nondiagnostic standard 12-lead ECG. Despite the value of the resting ECG, its sensitivity for ischemia is poor—as low as 20% in some studies.
Abnormalities of the ST segment and T wave may occur in a variety of conditions, including pulmonary embolism, ventricular hypertrophy, acute and chronic pericarditis, myocarditis, electrolyte imbalance, and metabolic disorders. Notably, hyperventilation associated with panic disorder can also lead to nonspecific ST and T-wave abnormalities. Pulmonary embolism is most often associated with sinus tachycardia but can also lead to rightward shift of the ECG axis, manifesting as an S-wave in lead I, with a Q-wave and T-wave in lead III (Chaps. 268 and 300). In patients with ST-segment elevation, the presence of diffuse lead involvement not corresponding to a specific coronary anatomic distribution and PR-segment depression can aid in distinguishing pericarditis from acute MI. CHEST RADIOGRAPHY
(See Chap. 308e) Plain radiography of the chest is performed routinely when patients present with acute chest discomfort and selectively when individuals who are being evaluated as outpatients have subacute or chronic pain. The chest radiograph is most useful for identifying pulmonary processes, such as pneumonia or pneumothorax. Findings are often unremarkable in patients with ACS, but pulmonary edema may be evident. Other specific findings include widening of the mediastinum in some patients with aortic dissection, Hampton’s hump or Westermark’s sign in patients with pulmonary embolism (Chaps. 300 and 308e), or pericardial calcification in chronic pericarditis. CARDIAC BIOMARKERS
Laboratory testing in patients with acute chest pain is focused on the detection of myocardial injury. Such injury can be detected by the presence of circulating proteins released from damaged myocardial cells. Owing to the time necessary for this release, initial biomarkers of injury may be in the normal range, even in patients with STEMI. Because of superior cardiac tissue-specificity compared with creatine kinase MB, cardiac troponin is the preferred biomarker for the diagnosis of MI and should be measured in all patients with suspected ACS at presentation and repeated in 3–6 h. Testing after 6 h is required only when there is uncertainty regarding the onset of pain or when stuttering symptoms have occurred. It is not necessary or advisable to measure troponin in patients without suspicion of ACS unless this test is being used specifically for risk stratification (e.g., in pulmonary embolism or heart failure).
The development of cardiac troponin assays with progressively greater analytical sensitivity has facilitated detection of substantially lower blood concentrations of troponin than was previously possible. This evolution permits earlier detection of myocardial injury, enhances the overall accuracy of a diagnosis of MI, and improves risk stratification in suspected ACS. The greater negative predictive value of a negative troponin result with current-generation assays is an advantage in the evaluation of chest pain in the ED. Rapid rule-out protocols that use serial testing and changes in troponin concentration over as short a period as 1–2 h appear promising and remain under investigation. However, with these advantages has come a trade-off: myocardial injury is detected in a larger proportion of patients who have non-ACS cardiopulmonary conditions than with previous, less sensitive assays. This evolution in testing for myocardial necrosis has rendered other aspects of the clinical evaluation critical to the practitioner’s determination of the probability that the symptoms represent ACS. In addition, observation of a change in cardiac troponin concentration between serial samples is useful in discriminating acute causes of myocardial injury from chronic elevation due to underlying structural heart disease, end-stage renal disease, or interfering antibodies. The diagnosis of MI is reserved for acute myocardial injury that is marked by a rising and/or falling pattern—with at least one value exceeding the 99th percentile reference limit—and that is caused by ischemia. Other non-ischemic insults, such as myocarditis, may result in myocardial injury but should not be labeled MI.
Other laboratory assessments may include the d-dimer test to aid in exclusion of pulmonary embolism (Chap. 300). Measurement of a B-type natriuretic peptide is useful when considered in conjunction with the clinical history and exam for the diagnosis of heart failure. B-type natriuretic peptides also provide prognostic information regarding patients with ACS and those with pulmonary embolism. Other putative biomarkers of acute myocardial ischemia or ACS, such as myeloperoxidase, have not been adopted in routine use. INTEGRATIVE DECISION-AIDS
Multiple clinical algorithms have been developed to aid in decision-making during the evaluation and disposition of patients with acute nontraumatic chest pain. Such decision-aids have been derived on the basis of their capacity to estimate either of two closely related but not identical probabilities: (1) the probability of a final diagnosis of ACS and (2) the probability of major cardiac events during short-term follow-up. Such decision-aids are used most commonly to identify patients with a low clinical probability of ACS who are candidates either for early provocative testing for ischemia or for discharge from the ED. Goldman and Lee developed one of the first such decision-aids, using only the ECG and risk indicators—hypotension, pulmonary rales, and known ischemic heart disease—to categorize patients into four risk categories ranging from a <1% to a >16% probability of a major cardiovascular complication. The Acute Cardiac Ischemia Time-Insensitive Predictive Instrument (ACI-TIPI) combines age, sex, chest pain presence, and ST-segment abnormalities to define a probability of ACS. More recently developed decision-aids are shown in Fig. 19-3. Elements common to each of these tools are (1) symptoms typical for ACS; (2) older age; (3) risk factors for or known atherosclerosis; (4) ischemic ECG abnormalities; and (5) elevated cardiac troponin levels. Although, because of very low specificity, the overall diagnostic performance of such decision-aids is poor (area under the receiver operating curve, 0.55–0.65), they can help identify patients with a very low probability of ACS (e.g., <1%). Nevertheless, no such decision-aid (or single clinical factor) is sufficiently sensitive and well validated to use as a sole tool for clinical decision-making.
Clinicians should differentiate between the algorithms discussed above and risk scores derived for stratification of prognosis (e.g., the TIMI and GRACE risk scores, Chap. 295) in patients who already have an established diagnosis of ACS. The latter risk scores were not designed to be used for diagnostic assessment. PROVOCATIVE TESTING FOR ISCHEMIA
Exercise electrocardiography (“stress testing”) is commonly employed for completion of risk stratification of patients who have undergone an initial evaluation that has not revealed a specific cause of chest discomfort and has identified them as being at low or selectively intermediate risk of ACS. Early exercise testing is safe in patients without high-risk findings after 8–12 h of observation and can assist in refining their prognostic assessment. For example, of low-risk patients who underwent exercise testing in the first 48 h after presentation, those without evidence of ischemia had a 2% rate of cardiac events through 6 months, whereas the rate was 15% among patients with either clear evidence of ischemia or an equivocal result. Patients who are unable to exercise may undergo pharmacological stress testing with either nuclear perfusion imaging or echocardiography. Notably, some experts have deemed the routine use of stress testing for low-risk patients unsupported by direct clinical evidence and a potentially unnecessary source of cost.
Professional society guidelines identify ongoing chest pain as a contraindication to stress testing. In selected patients with persistent pain and nondiagnostic ECG and biomarker data, resting myocardial perfusion images can be obtained; the absence of any perfusion abnormality substantially reduces the likelihood of coronary artery disease. In some centers, early myocardial perfusion imaging is performed as part of a routine strategy for evaluating patients at low or intermediate risk of ACS in parallel with other testing. Management of patients with normal perfusion images can be expedited with earlier discharge and outpatient stress testing, if indicated. Those with abnormal rest perfusion imaging, which cannot discriminate between old or new myocardial defects, must undergo additional in-hospital evaluation. OTHER NONINVASIVE STUDIES
Other noninvasive imaging studies of the chest can be used selectively to provide additional diagnostic and prognostic information on patients with chest discomfort. Echocardiography
Echocardiography is not necessarily routine in patients with chest discomfort. However, in patients with an uncertain diagnosis, particularly those with nondiagnostic ST elevation, ongoing symptoms, or hemodynamic instability, detection of abnormal regional wall motion provides evidence of possible ischemic dysfunction. Echocardiography is diagnostic in patients with mechanical complications of MI or in patients with pericardial tamponade. Transthoracic echocardiography is poorly sensitive for aortic dissection, although an intimal flap may sometimes be detected in the ascending aorta. CT Angiography
(See Chap. 270e) CT angiography is emerging as a modality for the evaluation of patients with acute chest discomfort. Coronary CT angiography is a sensitive technique for detection of obstructive coronary disease, particularly in the proximal third of the major epicardial coronary arteries. CT appears to enhance the speed to disposition of patients with a low-intermediate probability for ACS; its major strength being the negative predictive value of a finding of no significant disease. In addition, contrast-enhanced CT can detect focal areas of myocardial injury in the acute setting as decreased areas of enhancement. At the same time, CT angiography can exclude aortic dissection, pericardial effusion, and pulmonary embolism. Balancing factors in the consideration of the emerging role of coronary CT angiography in low-risk patients are radiation exposure and additional testing prompted by nondiagnostic abnormal results. MRI
(See Chap. 270e) Cardiac magnetic resonance (CMR) imaging is an evolving, versatile technique for structural and functional evaluation of the heart and the vasculature of the chest. CMR accurately measures ventricular dimensions and function and can be performed as a modality for pharmacologic stress perfusion imaging. Gadolinium-enhanced CMR can provide early detection of MI, defining areas of myocardial necrosis accurately, and can delineate patterns of myocardial disease that are often useful in discriminating ischemic from non-ischemic myocardial injury. Although usually not practical for the urgent evaluation of acute chest discomfort, CMR can be a useful modality for cardiac structural evaluation of patients with elevated cardiac troponin levels in the absence of definite coronary artery disease. CMR coronary angiography is in its early stages. MRI also permits highly accurate assessment for aortic dissection but is infrequently used as the first test because CT and transesophageal echocardiography are usually more practical.