Common Misdiagnoses of Myocardial Infarction

Remember how you were told in medical school that history is 90% of diagnosis? Was it usually ascribed to William Osler or another such legendary figure? Do you still believe it? I don't, or at least as not as much as I once did. What's changed? Well, we know more. We have more diagnostic techniques that enable us to ferret out the atypical presentations of disease and expose the limitations of clinical diagnosis. Did Sir Osler have to deal with a diagnosis made by MRA in a patient with vague symptoms?

This issue continues the discussion of missed MIs, focusing on how to distinguish infarction from other causes of chest pain. A continual challenge, especially with lethal disorders such as aortic dissection and pulmonary embolism in the differential.

I will close with another quote from Sir Osler: "There is no disease more conducive to clinical humility than aneurysm of the aorta." I can't say it any better than that.

—J. Stephan Stapczynski, MD, Editor

One of the problems facing emergency physicians attempting to diagnose every acute myocardial infarction (MI) patient is that the complaint of chest pain is both very common and results from a wide variety of causes. Further, while many causes of chest pain are benign and self-limited (reflux, anxiety, musculoskeletal pain), some causes can be emergent and deadly if untreated (aortic dissection, pulmonary embolism). Several studies that specifically analyzed patients with missed MI in the ED have identified some common misdiagnoses.1-4 For example, Pope, et al., listed the following discharge diagnoses for 19 patients with missed MI: 47% were diagnosed with non-cardiac chest pain, 16% with pulmonary problems, and 11% with stable angina.1 Examples of non-cardiac diagnoses include "atypical" chest pain (32%), musculoskeletal pain (21%), and gastrointestinal pain (5%). These conditions will be discussed briefly to identify strategies that may help discover patients with acute MI before they are discharged with one of these diagnoses. In addition, dangerous diseases that can be confused with acute MI, such as aortic dissection and pulmonary embolism (PE), also will be discussed.

Dangerous Conditions Confused with Acute MI

These patients are not among the missed MIs sent home, but they represent the other side of the coin: those who are admitted with misdiagnosis of MI. While these misdiagnoses occur less often than patients discharged with MI, it is useful to review these conditions briefly in the context of misdiagnosis of MI.

Thoracic Aortic Dissection. Compared to acute MIs, acute thoracic aortic dissections are rare. There were 1.2 million acute MIs presenting to EDs in the United States in 2008,5 but the average number of aortic dissections per year ranges from 2000–7000.6,7 This means that, on average, an ED physician will see up to 600 MI patients for every 1 aortic dissection. Risk factors are similar for both conditions (hypertension, cigarette smoking), and in both situations older patients present with acute chest pain. One recent review of 90 aortic dissection patients found that 67% developed acute dissection during exertion or emotional events,8 triggers traditionally associated more often with acute coronary syndrome (ACS) than with dissection. While most dissections occur in patients in their 60s and 70s, select younger patients also are at risk. Patients with bicuspid aortic valves, connective tissue disorders (Ehlers-Danlos), pregnant women, and crack cocaine users are at heightened risk.7 Ehlers-Danlos patients make up 5% of all aortic dissections, and half of all dissections in women younger than 40 years occur during the third trimester of pregnancy.9

Given that aortic dissection is relatively rare and can present in a multitude of various syndromes (acute CHF, MI, stroke, ischemic lower limb, paraplegia, gastrointestinal ischemia), it should not be surprising that accurate diagnosis remains elusive. One recent study of 66 patients with dissection found a misdiagnosis rate of 39%,10 a number that has been consistent since the 1980s.11 The most common misdiagnosis of aortic dissection in that study was acute coronary syndrome, also unchanged over the past 40 years.12 This occurs not just because patients are having chest pain, but their ECGs can be suggestive of acute ischemia. In the IRAD (international registry of aortic dissection) database, only 31% of patients with dissection had normal ECGs.13 ECG findings consistent with acute MI also are not rare in dissection.14 One study found up to 22% of acute dissections had ECG findings suggestive of non-STEMI and 4% had STEMI findings.15 The "needle in a haystack" concept is relevant. One author estimated that this translates into 25-200 cases of acute dissection with STEMI changes on the ECG in the entire United States annually, compared to the tens of thousands of patients with actual STEMIs.16 Patients with acute aortic regurgitation caused by the dissection will present with symptoms of acute CHF (dyspnea, pulmonary edema), and also may be misdiagnosed as such.

Besides misleading ECGs, the most specific cardiac enzyme for ischemia, troponin I, was found elevated in 23% of patients with aortic dissection as well.17 Given the shared risk factors for both aortic dissection and coronary artery diseases in these patients, undoubtedly some of the patients with elevated troponin actually are experiencing cardiac ischemia as well as aortic dissection.15 This further underscores the difficulty in accurately distinguishing the two conditions. The recent example of actor John Ritter, who died of a dissection at a California hospital on the way to the catheterization laboratory for presumed acute MI, emphasizes both how easy it is to confuse dissection with MI and how difficult it is for relatives to accept when "healthy" patients die suddenly in the ED.

Unfortunately misdiagnosing an aortic dissection as an acute MI can lead to disaster in several ways. Untreated mortality rates are quoted as 1% per hour7 for the first 24–48 hours. One recent study found that patients with acute dissection treated for MI instead had similar mortality rates to untreated patients, with a case fatality rate of 71% for misdiagnosed patients.10 The current standard treatment for acute MI involves several anticoagulants (heparin, aspirin, anti-platelet drugs) as well as thrombolytics. Administering these medications to patients with aortic dissection is associated with heightened risk via increased bleeding. In this study of patients misdiagnosed with ACS, 100% received aspirin, 85% heparin, and 12% fibrinolytic agents.10 The pressures on ED staff to reduce door-to-drug times18 coupled with the fact that hospitals often adopt pay-for-performance incentives to meet time-sensitive goals (antibiotics for pneumonia, aspirin for acute MI, etc.)19 means that it may be more difficult for ED physicians to pause in treatment of suspected acute MI patients and take the time to rule out a dissection.

The question, then, is what items can the ED physician use to help identify the patients who need to have aortic dissection excluded as a diagnosis? While aortic dissections are too variable for any guidelines to be highly specific, the following demographics, signs, and symptoms can be helpful. Much of the data below come from the international registry of aortic dissection (IRAD — see www.iradonline.org). First, the mean age for presentation is 63 years,13 and 68% of patients are men.20 Severe and sudden onset of pain may be the most useful clinical feature of all. Patients described their pain as severe or "worst ever" in 91% of cases, and pain came to full strength from the onset in 85%.13 Classic symptoms of "tearing" pain or migratory chest pain are more specific for dissection but occur less often. Only 50% described the pain as ripping or tearing, and pain was migratory in only 16%.13 However, 64% of patients described the pain as "sharp."13 Syncope in combination with acute chest pain is seen at similar rates in both MI and dissection (about 10%),21,22 but is seen more often with proximal dissection and may indicate development of cardiac tamponade.22 Thus, syncope with aortic dissection correlates with markedly increased mortality.16

Physical signs also can be suggestive of acute dissection. While hypertension is a well-established risk factor for aortic dissection, it is not consistently found on presentation. Of IRAD patients, only 35% of patients with proximal dissection had BP > 150 mmHg; however, 70% of distal dissections had hypertension.13 Hypotension is an ominous finding associated with a substantially increased risk of death.23 The most useful physical finding described is pulse deficits, defined as unilateral weakly palpated or absent pulses.7 These can be detected on physical exam, e.g. an absent left carotid pulse, or by noting significant differences (> 20 mmHg)24 comparing blood pressure of the extremities. Be aware that pulse deficits in patients with distal dissections may be present only in the legs and that pulse deficits may be transient.7 Roughly 15–30% of patients will have pulse deficits,13,25 thus stressing the importance of developing the habit of screening suspected acute MI patients for isolated absent pulses. When found, pulse deficits also correlate with higher in-hospital complications (coma, renal failure, limb ischemia) as well as higher mortality (44% vs. 25%).25

Diagnosis of acute aortic dissection can be confirmed only by one of four imaging studies: CT scan, TEE (transesophageal echocardiogram), angiography, or MRI. One may fortuitously see an intimal flap in the abdominal aorta using bedside ultrasound,26 but typically one of these four studies is required to make the diagnosis. Chest radiographs can be helpful when they show findings suspicious for aortic dissection (widened mediastinum, displaced calcification, tracheal displacement), but one recent review of 109 patients found only a 67% sensitivity for dissection.27 Plain film findings also were very non-specific. For example, more patients without dissection had tracheal deviation than those with the disease.27 The IRAD database shows that only 61% of patients had a widened mediastinum and that 12% of chest radiographs were normal.13

Attempts have been made to identify serum markers that can be used to screen patients with chest pain to pull out those with dissections.28 To date, the most accurate marker appears to be plasma D-dimer levels.29 D-dimer levels > 400 ng/mL were found in a study of 94 patients with acute dissection to be 99% sensitive but only 34% specific.29 Unfortunately, D-dimer levels are not precise enough to be used alone to rule out dissections; patients with thrombosed false lumens may be missed by D-dimer alone.30 Patients with any high-risk feature (sudden tearing pain, pulse deficits, widened mediastinum, etc.) in their presentation should continue to undergo definitive imaging (CT scan, TEE, angiography, or MRI).30

The bottom line at this time is that patients with chest pain atypical for dissection, ECGs suggestive of ischemia, and normal/nonspecific chest radiographs are more likely to be misdiagnosed or experience delayed diagnosis.31 Other features identified as correlating with a delay of diagnosis were pleural effusion on chest radiograph and patients presenting with respiratory distress.32 Overall, three high-risk features have been identified in retrospective analysis of patients with aortic dissection: sudden onset and/or tearing/ripping pain, widened mediastinum on chest radiograph, and pulse/blood pressure differences.24 Aortic dissection was found in only 7% of patients without all three features, but 83% had all three.24 One must remember these "classic" findings; unfortunately, cases where textbook presentations have been missed are not that rare.33

At centers with emergency angioplasty, the push to keep door-to-balloon times under 90 minutes can be beneficial for patients with dissection instead of MI. Patients taken rapidly to the catheterization laboratory from the ED with minimal work-up can have their dissection quickly identified by aortogram. The disadvantage is that they may be more likely to be given anticoagulants. Patients who are given tPA, such as those being transferred from smaller hospitals to those with intervention cardiology, are likely at higher risk from misdiagnosis. Clearly it is not constructive to obtain CT scans of every patient with an acute MI to rule out dissection. Besides the time delay, increased dye load from both CT scan and cardiac catheterization undoubtedly would cause more cases of acute renal failure than diagnosis of acute dissections. See Table 1 for suggestions on minimizing misdiagnosis.

Pulmonary Embolism. While acute pulmonary embolism (PE) may be confused less often with acute MI, there are certainly more acute PEs than aortic dissections, giving greater opportunity to misdiagnose PE as an MI. PE is the number three killer in the United States, with an estimated 50,000-100,000 deaths per year.34 It is difficult to obtain precise numbers, as many PEs are diagnosed only on autopsy and perhaps as many as 50% of these are not even suspected antemortum.35 On the other hand, the prevalence of PE is only 25-35% in patients who are evaluated by physicians for PE.36 Thus, the unhappy reality is that physicians are looking hard for PE largely in patients who do not have the condition, and "missing" it in those who do. One must wonder how so many PEs can go undetected if ED physicians are searching so diligently that up to 75% of patients are screened for PE do not have the diagnosis.

The answer may be that most of the PEs not detected by physicians may be clinically silent; thus, many PEs are not found because the patients with them are not presenting for medical evaluation. Studies in which VQ scans were done on all patients with proximal DVT found that 40-50% of them had silent PEs.37 While many of these silent PEs were silent for a reason (they were minor and had no clinical impact), nearly 6% of the silent PEs consisted of > 6 perfusion defects.37 Younger patients can obstruct up to 50% of their pulmonary tree, yet remain asymptomatic.38 Given the fact that a large percentage of DVTs are also silent (up to 80% of patients with PE have silent DVT),39 this implies that many patients are not presenting for care or are not developing symptoms that permit diagnosis. Thus, asymptomatic PEs are 4-5 times more common than symptomatic ones,40 leading at least one author to state, "There can only be a limited advantage to encourage increased alertness for a disease that is usually asymptomatic."41 However, this is not to undermine the importance of accurate diagnosis when patients do come to the ED with PE. Mortality from untreated PE can be 15-25% compared to 5% in treated patients.42

Multiple risk factors for PE have been identified (immobility, recent trauma/surgery, presence of cancer, hormone/birth control therapy, obesity, etc.), but one may not appreciate that elderly patients, especially those with chronic obstructive pulmonary disease (COPD) or congestive heart failure (CHF) also are at higher risk for PE.43 Be aware of acute dyspnea in these patients that differs from prior exacerbations. The overall mean age for presentation with a PE is 60 years, and PE occurs 10 times more often in patients older than 75 years compared to those younger than 40 years.44 Age may be an independent risk factor or it may just be that older people are more likely to have recent surgery and underlying disease states (i.e., lung cancer) that predispose them to DVT and PE. Regardless, given the fact the acute MI and PE are both especially difficult to diagnose in elderly patients, this population is likely most at risk for misdiagnosis in the ED.

Clinical presentation of PE tends to separate into three categories: massive PE with circulatory collapse, small PE with minimal to no symptoms, and moderate PE with variable degrees of chest pain and dyspnea.34 Patients with moderate PE may be at highest risk for misdiagnosis as acute MI. Their symptoms of dyspnea may be attributed to an anginal equivalent, and chest pain descriptions can be vague or described as "sharp," a term used by 20% of patients with PE45 and 22% of patients with acute MI.3 Be aware that syncope is a presenting symptom in up to 13% of patients with PE46 but is much less common for acute MI (< 1%).47,48

ECGs can be helpful when signs of acute right heart strain are present (S1Q3T3 or new RBBB), but these findings can be transient.49 Further, up to 27% of ECGs are normal in acute PE.50 The most common ECG findings of all are sinus tachycardia and non-specific ST-T wave changes,51 which are not useful for diagnosis. Several ECG patterns are known to occur with PE that can mimic acute MI. Right bundle branch block (RBBB) can be associated with ST segment elevation/depression and upright T waves in lead V1 and V2, thus simulating anterior or posterior MI patterns.52 More commonly, the ECG will show ST segment depression and inverted T waves in leads III, V2, and V3, a pattern that is not a typical anatomic distribution for acute MI. Other studies have show T wave inversions in inferior and anterior leads,52 which easily could simulate myocardial ischemia. However, one study looked specifically at differences in T wave inversion in patients with ACS and those with acute PE. They found that only 1% of patients with ACS had inverted T waves in leads III and V1, but these changes were seen in 88% of patients with PE.53 Although this study only included 40 patients with PE and 87 patients with ACS, these ECG differences could be helpful if validated in a larger group. Remember, though, that as with acute dissection, some of these patients actually may have cardiac ischemia as a complication of PE, making it even harder to render an accurate diagnosis.

Troponin levels also can be misleading. Several studies have noted that troponin I levels are elevated in 32%54 to 45%55 of patients with PE. Troponin levels tend to be mildly elevated and usually are less than 4-5 ng/dL.56 These patients tend to have moderate to massive PEs, and troponin levels appear to result from increased right heart strain. Elevated troponin levels in acute PE are associated with higher complication rates, cardiogenic shock, and increased mortality.56 D-dimer levels are known to be elevated in acute PE, but they are not specific enough and generally are limited to ruling out PE in low-risk patients only.43 PIOPED II shows at least 15% of patients with high suspicion of PE will have a negative D-dimer.57 Conversely, D-dimer levels also can be elevated in acute MI,58 but D-dimer levels have not been consistently shown to be helpful in diagnosing acute MI.59,60

The bottom line in these patients is to remember that mildly elevated troponin levels combined with T-wave inversions on the ECG do not necessarily diagnose the patient with an acute MI. T wave inversions that do not follow expected patterns for coronary ischemia, such as isolated inversions in leads III and V1, are more indicative of PE than MI. Be especially vigilant for PE when these findings occur in elderly patients with history of COPD or CHF. In addition, PIOPED II data show that chest CTs are negative in up to 18% of patients with high suspicion of PE.57 Table 2 provides a summary of tips for distinguishing acute MI from PE.

Benign Conditions Confused with Acute MI

GERD (Gastroesophageal Reflux Disease). (See Table 3.) The classic dilemma of differentiating pain of cardiac ischemia from that of heartburn shows how much the two conditions can overlap. Overall, indigestion is probably the most common misdiagnosis of STEMI patients.61 One study found that 58% of patients discharged after admission for possible ACS were found to have esophageal dysfunction instead.62 Reflux also is very common; as reported in one study, heartburn occurred in 7% of patients daily, 14% of patients weekly, and 15% of patients monthly.63 This totals 36% of patients experiencing reflux symptoms at least once a month, and it is estimated that 60% of adults will have at least one episode in their lifetime.64

It is easy to see why it can be hard to tell the two apart. Chest pain in GERD can be a dull burning sensation or can be ischemic-sounding with a squeezing chest pain associated with nausea, vomiting, and diaphoresis. One study found that squeezing chest pain was described as often as burning pain in patients with reflux.3 Radiation of pain to the neck, jaw, shoulders, back, and arms also is seen with GERD.64 Both chest pain from GERD and from angina have been reported as relieved by belching.65 Reflux also can be precipitated by exertion and relieved by rest or precipitated by emotional upset. Diabetes, obesity, and tobacco use are common risk factors for development of GERD, as they are associated with esophageal dysmotility. Abdominal pain also is a common chief complaint in acute MI patients who present without typical chest pain and is second only to dyspnea.66 Overall, however, abdominal pain is a relatively rare chief complaint of patients presenting with acute MI.67 When it does occur, though, it often is from inferior MIs in younger male patients, the same patient group that frequently complains of reflux symptoms.

Further complicating the picture, relief of symptoms with nitroglycerin is not helpful to identify anginal pain. In a study of 251 ED patients with chest pain, 88% of those with cardiac chest pain had some relief with sublingual nitroglycerin, but so did 92% of those with non-cardiac chest pain.68 Another study confirmed these numbers and again found more patients without cardiac chest pain had relief with nitroglycerin (41%) than those with coronary disease (35%).69 Likewise, relief with antacids is well described in both reflux and patients with proven myocardial ischemia.70 Be careful not to use the patient's response to these treatments as decision points in their evaluation.

Fortunately, there are a few features of GERD that, when present, can help distinguish it from angina. GERD more typically is associated with reflux of acid into the mouth and/or water brash, described as a hypersalivation that produces up to 10 mL in less than 1 minute.64 Reflux symptoms can be positional, meaning they are worse with bending over or lying down. They often are triggered by meals, especially larger meals, or spicy/hot foods. The fact that 36% of adults may have reflux symptoms monthly means that most patients with GERD are likely to have had similar symptoms chronically. Patients who have had reflux well documented in the past who describe their symptoms as such are less likely to be suffering from a new condition. Likewise, patients with a history of documented angina are more likely to describe their MI pain as a much more intense version (i.e., a similar location and feeling) of their previous angina. Identifying these components of the history underscores how important it can be for the ED physician to be thorough. Detecting these features can help support the diagnosis of GERD. Ultimately, history alone will not rule out MI in most patients.71 It must be emphasized that changes in a patient's reflux symptoms prompting a visit to the ED when he or she has not presented in the past should be taken seriously, and further evaluation likely is indicated.

Musculoskeletal Chest Pain. Chest pain resulting from musculoskeletal sources is a very common reason for patients to seek medical care. A large study of more than 24,000 outpatient visits found it to be the number one diagnosis in these patients, causing 49% of the visits, compared to only 12% being due to coronary artery disease.72 While chest pain that is pleuritic or reproduced by deep breaths is less likely to be cardiac in origin, this finding alone does not rule it out.73 Older studies have reported that pleuritic pain was seen in 15% of patients with acute MI.74 More recent ones have found that a similar number of patients with acute MI (13%) did have a component of their pain described as pleuritic, but none of the patients whose pain was entirely pleuritic had active coronary disease.73

Reproducibility of the patient's pain by palpation of the chest wall has been presented in the past as strong evidence that the pain's source is not due to coronary disease.75 This aspect of the physical examination also is used in well known protocols used to identify patients with acute MI in ED patients.76 More recent studies have confirmed that a small percentage of patients with acute MI will have this feature (7-15%),73,74 but this number remains high enough that one should not use reproducibility of pain by palpation as a reliable single feature to rule out possible MI.

Anxiety. Anxiety disorders are familiar to seasoned ED physicians and are some of the most common psychiatric disorders. It is estimated that up to 25% of the U.S. population suffers from pathologic anxiety during their lifetime.77 In the ED, one must be careful when attempting to diagnose a new-onset anxiety disorder. Up to 42% of patients presenting with apparent anxiety disorders are later found to have organic disease instead.78 Conversely, anxiety disorders are seen in up to 30% of patients initially presenting with chest pain who later are diagnosed as non-cardiac chest pain.79

Symptoms of anxiety include hyperventilation, feeling short of breath, palpitations, chest pain — complaints that can be consistent with acute MI as well. Patients presenting with an anxiety attack feel impending doom and experience intense hyperventilation. When these patients roll into the ED, it often takes some serious thought as to whether they may be right when they breathlessly state "I'm going to die!" Indeed, dyspnea is a common complaint in acute MI patients, especially elderly ones. Dyspnea has been shown to be the single most common complaint for acute MI in patients older than 85 years.80 In older studies, dyspnea is present in 33% of patients with acute MI,81 but more recent work still shows dyspnea in 14% of acute MIs and in 5% of patients with unstable angina.67 However, another study of unstable angina found that dyspnea was seen in 69% of patients.82

In general, anxiety disorders tend to occur in females twice as often as males, and the average age of onset is in the third decade. Unfortunately, this may only confuse the picture, as younger patients and women are among the groups where MI most often is missed. When treating acute anxiety in the ED in patients where the diagnosis is not clear, further evaluation to screen for acute MI often is needed.83

The Patient with Recent Stress or Catheterization

When patients present to the ED with chest pain and a history of known coronary disease, the decision of whether to admit the patient can be challenging. While it may seem on the surface that these patients would be an automatic admission, in real life this is not always what happens. The patient is not always experiencing another acute MI or unstable angina, and if it is safe to do so, then the patient should be discharged. The ED physician must risk stratify these patients further into those who truly need admission and those who can be safely discharged. Many factors go into this decision, but this section will focus on the patient's previous cardiac testing (stress tests and cardiac catheterization) and the limits of these tests. Specifically, if a patient has had a recent catheterization (< 6-12 months) that is either normal or stable from previous studies, does this information help one to decide his or her disposition? The same question using recent stress tests also will be addressed. Likewise if the patient has had a recent negative stress or catheterization, how useful is this information?

At least two recent studies have addressed this issue, and the results were somewhat surprising in that history of a recent negative work-up did not appear to influence subsequent ED evaluations.84,85 One study looked at the utilization of resources in the year following an admission for chest pain that resulted in a non-cardiac diagnosis.85 Researchers followed 556 patients admitted for chest pain, 116 of whom had a negative stress test and 20 of whom had negative cardiac catheterizations. There was no difference in use of resources in the next year, defined as echocardiograms, stress tests, or catheterizations, between the study patients or controls.85 Another study examined patients with any history of negative stress tests and looked for an influence on the ED physician's decision to admit the patient.84 They also found no impact on admission rates but did not discriminate between patients with recent stress tests and those whose studies were several years old. As expected, patients with a previous abnormal stress test were admitted more frequently.84

In addition, as observation units or chest pain units are being used in increasing numbers for risk stratification of chest pain patients in the ED, emergency physicians are at times in charge of choosing which stress test to perform and managing the results. Therefore, overall limits of stress testing will be discussed as well. Ultimately, one should remember that if the patient has been evaluated recently by a cardiologist, one can always consult him or her for help in deciding difficult dispositions.

Recent Stress Test. Currently, most stress tests performed have two components: the classic treadmill or exercise part and the subsequent nuclear perfusion study. The treadmill test is performed by attaching an ECG monitor to a patient using a treadmill or stationary bicycle. If the patient is physically unable to walk on a treadmill or ride the bike, medications can be used to increase their heart rate instead (persantine or dobutamine). The most common protocol is the Bruce protocol, where resistance is increased in 3-minute intervals86 until the patient reaches a predetermined level, such as a maximum metabolic output or through physiological signs of stress (drop in blood pressure, angina, drop in skin temperature, or cyanosis). The ECG is monitored during exercise and recovery to identify ST segment depression or elevation indicating ischemic changes. Contraindications for exercise testing are listed in Table 4. The second part is the perfusion study, which is performed by injection of an isotope (thalium-201, or technetium-99 sestamibi) that allows one to assess cardiac perfusion during and after exercise. One can compare the exercise and rest images to identify areas of the heart that are not perfused normally only during exercise (reversible ischemia) or not at all (scarring).

As with any test, certain factors can make the test difficult to interpret or reduce its accuracy. For the standard exercise component, patients must have relatively normal baseline ECGs from which one can interpret transient ischemic changes. Patients with left ventricular hypertrophy or digitalis therapy often have baseline ST segment depression, making ECG changes harder to detect. Other conditions commonly associated with false-positive treadmill test results include valvular heart disease, anemia, interventricular conduction delays (including bundle branch blocks), hypokalemia, and significant hypoxia or volume overload.87 Results are reported in general as negative or positive for ECG changes suggestive of ischemia.

Overall, exercise testing is most useful when performed in populations who are at intermediate risk of coronary disease.88 The positive predictive value is high (98%) in patients suspected to have coronary disease and, likewise, the negative predictive value is also high (98%) in patients unlikely to have coronary disease.88 Thus an exercise test alone does little in screening very high risk or very low risk patients. It is best used in intermediate populations where the overall sensitivity and specificity are 68% and 77%, respectively.89 In general, contraindications for stress testing of emergency department patients include those with evolving ECG changes, abnormal cardiac biomarkers, worsening or persistent chest pain, and clinical risk profile that suggests coronary angiography is needed.87

In most hospitals, the exercise test is performed in conjunction with nuclear perfusion imaging studies (stress myoview, stress thallium). Images taken during and after exercise show general blood flow to the heart, and areas that do not increase flow during exercise indicate possible ischemia. Images are obtained using single-photon emission computed tomography (SPECT). Again, detecting areas of reversible ischemia indicate regions at risk for future MI. The overall sensitivity and specificity of perfusion studies is 90% and 70%, respectively.90 Since the study relies on comparison of muscle perfusion with the adjacent "normal" areas, perfusion studies have some inherent inaccuracy. False-negative results are seen in patients with overall low perfusion, such as in left main disease or proximal 3-vessel disease. They can be missed as the entire heart is not well perfused and no single area will stand out as different. Also not all perfusion differences are caused by large-vessel coronary disease. Patients with microvascular disease (diabetes, hypertension, and hyperlipidemia) may have significant flow imbalances without obstructive large-vessel lesions. Breast implants also can make the study more difficult to interpret.

Documentation of a recent perfusion study can be more helpful with patient disposition than a simple exercise test. The extent of the perfusion defect directly correlates to the patient's risk of MI or death. Patients with normal perfusion studies have only a 0.4% risk of MI/death per year compared to a 9% risk for those with multiple abnormal segments.91 Large studies (21,000 patients) have confirmed these numbers, and one 6-year follow-up survey still found an annual rate of MI/death of 0.88%.92 Even patients with known coronary disease have a low annual risk (0.9%) of MI/death when the perfusion study shows no reversible ischemia. Thus current cardiology texts state that a normal perfusion study can be used to predict a low risk of MI/death for as far as two years into the future (the "warranty period").93 Keep in mind, though, that the specificity and sensitivities quoted above are for the detection of coronary lesions > 50%, as this is the cut-off for the definition of coronary artery disease in most cardiology literature; some even use a level of > 70%.93 Thus, exercise tests and perfusion studies do not rule out the presence of coronary disease; they simply identify patients at risk of future events (i.e., those likely to have significant obstructing lesions).

Recent Cardiac Catheterization (< 6–12 months). The first cardiac catheterization on a living patient occurred in 1929 in Germany, when a surgical apprentice performed the procedure on himself. Even though Dr. Forssmann was later awarded the Nobel Prize for his work in 1956, at the time, he was fired for doing the unauthorized procedure. Selective coronary angiography began in 1958, and the first balloon angioplasty was done in 1974. In 1997, there were roughly 1 million cardiac catheterizations in the United States alone, and this number has climbed to 2.5-3 million in 2008.94 Besides defining coronary disease, a cardiac catheterization provides information on left ventricular dysfunction, measures pressures and cardiac output, and can assess valvular problems.

Multiple factors go into assessing a patient's risk using catheterization results. Patients with coronary lesions generally will be classed as having single-, double-, or three-vessel disease. When patients with coronary artery disease undergo catheterization, approximately 25% will have either single-, double-, or triple-vessel disease, defined as at least a 70% narrowing.95 Another 5–10% will have left main disease, and the remaining 15% will have no critical lesions (> 70% stenosis). Statistically, the more vessels involved, the higher the risk of future MI. The 5-year survival rate for patients with triple-vessel disease and a 95% proximal LAD lesion is 59%, compared to 93% for a patient with only low-grade single-vessel disease when both are treated medically.96 Even asymptomatic patients with triple-vessel disease experience 4-5% annual mortality.97

Patients with documented coronary artery disease on a previous catheterization who did not require intervention at the time are obviously at risk of future events. Most cardiologists only place stents in lesions with least 70% or greater stenosis, but they may stent a 50–70% lesion if the patient is experiencing unstable angina due to a particular lesion. The decision to place a stent can be complicated, as lesions are believed to need to be at least 95% or greater to be responsible for most rest angina, but only 15% of acute MIs arise from lesions > 60%.98 Thus, many patients will have previous catheterization reports documenting lesions less than 50% who are being medically managed.

Ultimately, one either may admit patients with chest pain and history of coronary disease, place them in a chest pain unit setting, or discharge them after serial biomarkers to rule out infarction. Patients who have a "good story" for their angina or who rarely seek care for their chest pain usually are better off being admitted negative biomarkers or not, and patients who are able to be discharged usually are those who are evaluated often for chest pain that turns out not to be MI or unstable angina or who are well known to their cardiologist. Discussion with the patient's cardiologist is advised before discharging for any help with the decision and to ensure timely follow up.

Recent Stent Placement. Coronary stents were introduced in the late 1990s as a solution to the high rate of restenosis (30-40% in only 6 months) seen in balloon angioplasty alone.99 Originally, bare metal stents were used, but these were associated with a 20-30% restenosis rate in 6-9 months due to intimal hyperplasia99 and caused angina in up to 15% of patients in 1 year after deployment.100 Subsequently, drug-eluting stents were developed in 2001 using agents released by the stent to reduce the extent of intimal hyperplasia.101 Drug-eluting stents (DES) have been successful in reducing the rate of restenosis down to only 4-6% and, as a result, 90% of all stents placed in the United States and Europe now are DES. However, patients given DES must be on a much longer regimen of aspirin and clopidogrel (Plavix) of up to 6–12 months or longer.101 Even after this treatment, they also must remain on an aspirin regimen for life.102 Current work with bioresolvable stents is likely to be the next development and may help improve outcomes even further.103

ED physicians need to be aware that certain groups of patients are at increased risk of complication after stent placement. (See Table 5.) Patients with a history of diabetes are at higher risk for restenosis leading to re-intervention, with double the rates of non-diabetics.104 Those who stop taking their aspirin and clopidogrel after stenting also are at increased risk for late-stent thrombosis, defined as occurring more than 30 days after stenting.101 Late-stent thrombosis is a dangerous event and is associated with a 45% mortality rate.105 Patients who have had stent placement in the last several months should be specifically asked if they are taking their clopidogrel, as they are at much higher risk for acute MI if they have stopped. All patients with DES also should be asked to verify that they have been compliant with their aspirin therapy. Patients in these higher risk categories should be discussed with their cardiologist.

The "Negative" Catheterization Result. Unfortunately, when the patient's history reveals negative catheterization results, one still must proceed with caution. (See Table 5.) First, beware of the patient's version of the catheterization results. One should verify that the results truly were negative for coronary lesions, as many times when the patient says "negative" it may be that coronary disease was indeed found but no intervention/stenting was performed. In other words, negative to a physician means no coronary artery disease was found, but to the patient, it may mean no action was required that day for the 40% proximal LAD lesion.

Secondly, a truly negative catheterization result does not mean the patient could not be having an acute MI now. Coronary spasm easily can cause a fatal STEMI and can occur in multiple ways. Prinzmetal angina was first described by Dr. Prinzmetal in 1959 to explain a syndrome in which patients had normal exercise tolerances but developed acute MI/angina at rest, often in the early morning hours.106 Coronary vasospasm is thought to be due to dysfunctional coronary endothelium or release of local vasoconstrictors from platelets (serotonin or thromboxane A2).107 Cocaine also is a well-established cause of coronary vasospasm.

Myocardial bridging is another mechanism by which MI still can occur in patients with negative catheterization. This term refers to a large coronary vessel that usually is epicardial but that goes under the myocardial surface for short segments. This can cause physical compression of a coronary vessel by the surrounding muscle during systole. While seen in only about 5% of patients with normal vessels,95 the incidence in autopsy series is quoted between 15–85%.108 Bridging most often occurs in the middle segment of the LAD. A recent IVUS study suggested that the phenomenon is more frequent after all. The study found an incidence of 23%, but this was in a patient population with a high incidence of artery disease, suggesting that most bridging lesions are angiographically silent.109

Negative catheterization results are seen in up to 20% of patients referred for clinical suspicion of angina.95 Far more women (50%) than men (17%) with suspected angina have negative results or lesions < 50%.110 Even 10% of women and 6% of men who present with STEMI have no disease or only non-obstructing lesions.111 Some of these patients who have typical angina-sounding symptoms, multiple coronary risk factors (hypertension, diabetes, elevated cholesterol, abdominal obesity), and ST segment depression on exercise ECGs are referred to by cardiologists as "cardiac syndrome X."112 The source of the patients' pain has been debated in the past, but newer studies suggest they are experiencing subendocardial ischemia.113 This occurs not from large-vessel obstructions, but from very small vessel flow limitation or microvascular angina.95 In the past, prognosis of these patients was thought to be excellent. At least one recent study found that in patients in whom microvascular angina could be demonstrated, 14% had either cardiac death, MI, or CABG in a 28-month period.114

Finally, patients who have significant left ventricular hypertrophy also can develop mild, diffuse subendocardial ischemia. Essentially, the heart muscle becomes thick enough that the blood supply to the subendocardium cannot keep pace with demand. In times of increased demand, small areas can infarct but subsequent catheterization will reveal no blockage.

The bottom line is that most patients who have had truly negative catheterization in the recent past (< 6-12 months) are less likely to be infarcting than someone whose arteries have not been studied. However, the above scenarios do happen, and a negative catheterization alone does not rule out the possibility of an acute MI. Anyone who has multiple risk factors, especially diabetes and smoking history, who presents with a "good story" for angina still should be evaluated for possible acute MI or unstable angina. Overall, patients with negative catheterization reports have a low mortality rate. A study of survival rates for patients with normal catheterizations found that 96% were alive after 7 years, compared to 92% of patients who had mild coronary lesions (< 50%).112

References

1. Pope JH, Aufderheide TP, Ruthazer R, et al. Missed diagnoses of acute cardiac ischemia in the ED. N Engl J Med 2000;342:1163-1170.

2. Lee TH, Goldman L. Evaluation of the patient with acute chest pain. N Engl J Med 2000;342:1187-1195.

3. Schor S, Behar S, Modan B, et al. Disposition of presumed coronary patients from an emergency room. A follow-up study. JAMA 1976;236:941-943.

4. McCarthy BD, Beshansky JR, D'Agostino RB, et al. Missed diagnoses of acute myocardial infarction in the emergency department: Results from a multicenter study. Ann Emerg Med 1993;22:579-582.

5. Rosamond W, Flegal K, Furie K, et al. Heart disease and stroke statistics—2008 update: A report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation 2008;117:e25-146.

6. Klompas M. Does this patient have an acute thoracic aortic dissection? JAMA 2002;287:2262-2272.

7. Isselbacher EM. Diseases of the aorta. In: Libby P, Bonow RO, Mann DL, et al, eds. Braunwald's Heart Disease: A Textbook of Cardiovascular Medicine, 8th ed. Philadelphia, PA: Elsevier; 2008: 1457-1487.

8. Hatzaras IS, Bible JE, Koullias GJ, et al. Role of exertion or emotion as inciting events for acute aortic dissection. Am J Cardiol 2007;100:1470-1472.

9. Williams GM, Goh VL, Brawley RK, et al. Aortic disease associated with pregnancy. J Vasc Surg 1988;8:470-475.

10. Hansen MS, Nogareda GJ, Hutchison SJ. Frequency and inappropriate treatment of misdiagnosis of acute aortic dissection. Am J Cardiol 2007;99:852-856.

11. Spittal PC, Spittal JA Jr, Joyce JW, et al. Clinical features and differential diagnosis of aortic dissection: Experience with 236 cases (1980 through 1990). May Clin Proc 1993;68:642-651.

12. Viljanen T. Diagnostic difficulties in aortic dissection. Ann Chir Gynaecol 1986;75: 328-332.

13. Hagan PG, Nienaber CA, Isselbacher EM, et al. The international registry of acute aortic dissection (IRAD): New insights into an old disease. JAMA 2000;283;897-903.

14. Hirata K, Kyushima M, Asato H. Electrocardiographic abnormalities in patients with acute aortic dissection. Am J Cardiol 1995;76:1207-1212.

15. Biagini E, Lofiego C, Ferlito M, et al. Frequency, determinants, and clinical relevance of acute coronary syndrome-like electrocardiographic findings in patients with acute aortic syndrome. Am J Cardiol 2007;100:1013-1019.

16. Haro LH, Krajicek M, Lobl JK. Challenges, controversies, and advances in aortic catastrophes. Emerg Med Clin North Am 2005;23:1159-1177.

17. Bonnefoy E. Significant of serum troponin I elevation in patients with acute aortic dissection of the ascending aorta. Acta Cardiol 2005;60:165-170.

18. Bradley EH, Herrin J, Wang Y, et al. Door-to-drug and door-to-balloon times: Where can we improve? Time to reperfusion therapy in patients with ST segment elevation myocardial infarction (STEMI). Am Heart J 2006;151:1281-1287.

19. Sikka R. Pay for performance in emergency medicine. Ann Emerg Med 2007;49: 756-761.

20. Nienaber CA, Fattori R, Mehta RH, et al. Gender-related differences in acute aortic dissection. Circulation 2004;109: 3014-3021.

21. Canto J, Fincher C, Kiefe C, et al. Atypical presentations among Medicare beneficiaries with unstable angina pectoris. Am J Cardiol 2003;91:118-119.

22. Nallamothu BK, Metha RH, Saint S, et al. Syncope in acute aortic dissection: Diagnostic prognostic, and clinical implications. Am J Med 2002;113:468-471.

23. Tsai TT, Bossone E, Isselbacher EM, et al. Clinical characteristics of hypotension in patients with acute aortic dissection. Am J Cardiol 2005;95:48-52.

24. von Kodolitsch Y, Schwartz AG, Nienaber CA. Clinical prediction of acute aortic dissection. Arch Intern Med 2000;160: 2977-2982.

25. Bossone E, Rampoldi V, Nienaber CA, et al. Usefulness of pulse deficit to predict in-hospital complications and mortality in patients with acute type A aortic dissection. Am J Cardiol 2002;89:851-855.

26. Fojtik JP, Costantino TG, Dean AJ. The diagnosis of aortic dissection by emergency medicine ultrasound. J Emerg Med 2007; 32:191-196.

27. von Kodolitsch Y, Nienaber CA, Dieckmann C, et al. Chest radiography for the diagnosis of acute aortic syndrome. Am J Med 204;116:73-77.

28. Mir MA. Aortic dissection — in pursuit of a serum marker. Am J Emerg Med 2008;26:942-945.

29. Ohlmann P, Faure A, Morel O, el al. Diagnostic and prognostic value of circulating D-dimers in patients with acute aortic dissection. Crit Care Med 2006;34: 1358-1364.

30. Sutherland A, Escano J, Coon TP. D-dimer as the sole screening test for acute aortic dissection: A review of the literature. Ann Emerg Med 2008;52:339-343.

31. Raghupathy A, Nienaaber CA, Harris KM, et al. Geographic differences in clinical presentation, treatment, and outcome in type A acute aortic dissection (from the IRAD). Am J Cardiol 2008;102: 1562-1566.

32. Rapezzi C, Longhi S, Graziosi M, et al. Risk factors for diagnostic delay in acute aortic dissection. Am J Cardiol 2008; 102:1399-1406.

33. Rogers RL, McCormack R. Aortic disasters. Emerg Med Clin North Am 2004;22: 887-908.

34. Goldhaber SZ. Pulmonary embolism. In: Libby P, Bonow RO, Mann DL, et al, eds. Braunwald's Heard Disease: A Textbook of Cardiovascular Medicine, 8th ed. Philadelphia, PA: Elsevier; 2008: 1863-1879.

35. Pineda LA, Hathwar VS, Grant BJ. Clinical suspicion of fatal pulmonary embolism. Chest 2001;120:791-795.

36. PIOPED Investigators. Value of the ventilation/perfusion scan in acute pulmonary embolism: Results of the prospective investigation of pulmonary embolism diagnosis (PIOPED). JAMA 1990;263:2753-2759.

37. Meignan M, Rosso J, Gauthier H, et al. Systematic lung scans reveal a high frequency of silent PE in patients with proximal DVT. Arch Intern Med 2000; 160:159-164.

38. Riedel M. Diagnosing pulmonary embolism. Postgrad Med J 2004;80: 309-319.

39. Buller H, Sohne M, Middeldorp S. Treatment of venous thromboembolism. J Thromb Haemost 2005;3:1554-1560.

40. Egermayer P, Town GI. The clinical significance of pulmonary embolism: Uncertainties and implications for treatment—a debate. J Intern Med 1997;241: 5-10.

41. Egermayer P. Silent pulmonary embolism. Arch Intern Med 2000;160:2218.

42. Stein PD, Henry JW, Relyea B. Untreated patients with pulmonary embolism: Outcome, clinical, and laboratory assessment. Chest 1995;107:931-935.

43. Laack TA, Goyal DG. Pulmonary embolism: An unsuspected killer. Emerg Med Clin North Am 2004;22:961-983.

44. Kim V, Spandorfer J. Epidemiology of venous thromboembolic disease. Emerg Med Clin North Am 2001;19:839-859.

45. Stein PD, Beemath A, Matta F, et al. Clinical characteristics of patients with acute pulmonary embolism: Data from PIOPED II. Am J Med 2007;120:871-879.

46. Sarasin FP, Louis-Simonet M, Carballo D, et al. Prospective evaluation of patients with syncope: A population-based study. Am J Med 2001;111:177-184.

47. Link MS, Lauer EP, Homoud MK, et al. Low yield of rule-out MI protocol in patients presenting with syncope. Am J Cardiol 2001;88:706-707.

48. Brignole M, Alboni P, Benditt DG, et al. Guidelines on the management (diagnosis and treatment) of syncope — update. Eur Heart 2004;25:2054-2072.

49. Chan TC, Vilke GM, Pollack M, et al. Electrocardiographic manifestations: pulmonary embolism. J Emerg Med 2001; 21:263-270.

50. Panos RJ, Barish RA, Whye DW Jr, et al. The electrocardiographic manifestations of pulmonary embolism. Emerg Med 1988;6:301-307.

51. Ullman E, Brady WJ, Perron AD, et al. Electrocardiographic manifestations of PE. Am J Emerg Med 2001;19:514-519.

52. Sreeram N, Cheriex EC, Smeets JLRM, et al. Value of the 12-lead EKG at hospital admission in the diagnosis of PE. Am Cardiol 1994;73:298-303.

53. Kosuge M, Kimura K, Ishikawa T, et al. Electrocardiographic differentiation between acute PE and ACS on the basis of negative T waves. Am J Cardiol 2007; 99:817-821.

54. Giannitsis E, Müller-Bardorff M, Kurowski V, et al. Independent prognostic value of cardiac troponin T in patients with confirmed pulmonary embolism. Circulation 2000;102:211-217.

55. Aksay E, Yanturali S, Kiyan S. Can elevated troponin I levels predict complicated clinical course and inhospital mortality in patients with acute pulmonary embolism? J Emerg Med 2007;25:138-143.

56. Mehta NJ, Jani K, Khan IA. Clinical usefulness and prognostic value of elevated cardiac troponin I levels in acute PE. Am Heart J 2003;145:821-825.

57. Stein PD, Fowler SE, Goodman LR, et al; PIOPED II Investigators. Multidetector computed tomography for acute pulmonary embolism. N Engl J Med 2006;354: 2317-2327.

58. Sadosty AT, Boie ET, Stead LG. Pulmonary embolism. Emerg Med Clin North Am 2003;21:363-384.

59. Bayes-Genis A, Mateo J, Santalo M, et al. D-dimer is an early diagnostic marker of coronary ischemia in patients with chest pain. Am Heart J 2000;140:379-384.

60. Shitrit D, Bar-Gil Shitrit A, Rudensky B, et al. Determinants of ELISA D-dimer sensitivity for unstable angina pectoris as defined by coronary catheterization. Am J Hematol 2004;76:121-125.

61. Antman EM, Braunwald E. ST-elevation myocardial infarction: Pathology, pathophysiology and clinical features. In: Libby P, Bonow RO, Mann DL, et al, eds. Braunwald's Heart Disease: A Textbook of Cardiovascular Medicine, 8th ed. Philadelphia, PA: Elsevier; 2008: 1207-1230.

62. Areskog M, Tibbling L, Wranne B. Oesophageal dysfunction in non-infarction coronary care unit patients. Acta Med Scand 1979;205:279-282.

63. Nebel OT, Frones MF, Castell DO. Symptomatic gastroesophageal reflux: Incidence and precipitating factors. Am J Dig Dis 1976;21:953-956.

64. Lowell MJ. Esophagus, stomach and duodenum. In: Marx, JA ed. Rosen's Emergency Medicine: Concepts and Clinical Practice, 6th ed. St. Louis: Mosby; 2006: 1386-1398.

65. Pope JH, Selker HP. Acute coronary syndromes in the emergency department: Diagnostic characteristics, test, and challenges. Cardiol Clin 2005;23:423-451.

66. Clark LT, Adams-Campbell LL, Maw M, et al. Clinical features of patients with acute myocardial infarction presenting with and without typical chest pain: An inner city experience. J Assoc Acad Minor Phys 1989;1:29-31.

67. Pope J, Ruthazer R, Beshansky J, et al. Clinical features of emergency department patients presenting with symptoms of acute cardiac ischemia: A multicenter study. J Thromb Thrombolysis 1998;6:63-74.

68. Shry EA, Dacus J, Van De Graaff E, et al. Usefulness of the response to sublingual nitroglycerin as a predictor of ischemic chest pain in the emergency department. Am J Cardiol 2002;90:1264-1266.

69. Henrikson CA, Howell EE, Bush DE, et al. Chest pain relief with nitroglycerin does not predict active coronary artery disease. Ann Intern Med 2003;139:979-986.

70. Dobrzycki S, Baniukiewicz A, Korecki J, et al. Does gastro-esophageal reflux provoke myocardial ischemia in patients with CAD? Int J Cardiol 2005;104:67-72.

71. Swap CJ. Value and limitations of chest pain history in the evaluation of patients with suspected acute coronary syndromes. JAMA 2005;294:2623-2629.

72. Verdon F. Chest pain in daily practice: Occurrence, causes and management. Swiss Med Wkly 2008;138:340-347.

73. Lee TH, Cook EF, Weisberg M, et al. Acute chest pain in the emergency room: Identification and examination of low risk patients. Arch Intern Med 1985;145:65-69.

74. McElroy JB. Angina pectoris with coexisting musculoskeletal chest pain. Am Heart J 1963;66:96-99.

75. Panju AA, Hemmelgarn BR, Guyatt GH, et al. The rational clinical examination. Is this patient having a myocardial infarction? JAMA 1998;280:1256-1263.

76. Goldman L, Weinberg M, Weisberg M, et al. A computer-derived protocol to aid in the diagnosis of emergency room patients with acute chest pain. N Engl J Med 1982;307:588-596.

77. Kessler RC, KA McGonagle, S Zhao, et al. Lifetime and twelve-month prevalence of DSM III-R psychiatric disorders in the United States: Results from the National Comorbidity Survey. Arch Gen Psychiatry 1994;51:8-19.

78. Kercher EE. Anxiety. Emerg Med Clin North Am 1991;9:161-187.

79. Kanton WJ. Chest pain, cardiac disease and panic disorder. J Clin Psychiatry 1990; 51:27-30.

80. Jones ID, Slovis CM. Emergency department evaluation of the chest pain patient. Emerg Med Clin North Am 2001;19: 269-282.

81. Alonzo A, Simon A, Feilieb M. Prodromata of myocardial infarction and sudden death. Circulation 1975;52:1056-1062.

82. Canto JG, Shilpack MG, Rogers WJ, et al. Prevalence, clinical characteristics, and mortality among patients with myocardial infarction presenting without chest pain. JAMA 2000;283:3223-3229.

83. Kercher EE, Tobias JL. Anxiety disorders. In: Marx, JA ed. Rosen's Emergency Medicine: Concepts and Clinical Practice, 6th ed. St. Louis: Mosby; 2006: 1744-1752.

84. Nerenberg RH, Frances SS, Robey, et al. Impact of a negative prior stress test on emergency physician disposition decision in ED patients with chest pain syndromes. Am J Emerg Med 2007;25:39-44.

85. Shaver KJ, Marsan RJ, Sease KL, et al. Impact of a negative evaluation for underlying coronary artery disease on one-year resource utilization for patients admitted with potential acute coronary syndromes. Acad Emerg Med 2004;11:1272-1277.

86. Bruce RA. Exercise testing of patients with coronary artery disease. Ann Clin Res 1971;3:323-330.

87. Chaitman BR. Exercise stress testing. In: Libby P, Bonow RO, Mann DL, et al, eds. Braunwald's Heard Disease: A Textbook of Cardiovascular Medicine, 8th ed. Philadelphia, PA: Elsevier; 2008: 195-220.

88. Murthy TH, Bach DS. Comparative review of stress tests. Clin Fam Pract 2001;3: 801-816.

89. Gianrossi R, Detrano R, Mulvihill D, et al. Exercise-induced ST depression in the diagnosis of coronary artery disease: A meta-analysis. Circulation 1989;80:87-98.

90. Gibbons RJ, Chatterjee K, Daley J, et al. ACC/AHA/ACP-ASIM Guidelines for the Management of Patients with Chronic Stable Angina: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 1999;33:2092-2197.

91. Hachamovitch R, Berman DS, Kiat H, et al. Exercise myocardial perfusion SPECT in patients without known coronary artery disease: Incremental prognostic value and use in risk stratification. Circulation 1996;93:905-914.

92. Vanzetto G, Ormezzano O, Fagret D, et al. Long-term additive prognostic value of thallium-201 myocardial perfusion imaging over clinical and exercise stress test in low to intermediate risk patients: Study in 1137 patients with 6-year follow-up. Circulation 1999;100:1521-1527.

93. Udelson JE, Dilsizian V, Bonow RO. Nuclear cardiology. In: Libby P, Bonow RO, Mann DL, et al, eds. Braunwald's Heard Disease: A Textbook of Cardiovascular Medicine, 8th ed. Philadelphia, PA: Elsevier; 2008: 345-389.

94. Thom T, Haase N, Rosamond W, et al. Heart disease and stroke statistics-2006 update: A report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation 2006;113:e85-e151.

95. Morrow DA, Gershe BJ. Chronic coronary artery disease. In: Libby P, Bonow RO, Mann DL, et al, eds. Braunwald's Heard Disease: A Textbook of Cardiovascular Medicine, 8th ed. Philadelphia, PA: Elsevier; 2008: 1353-1405.

96. Califf RM, Armstrong PW, Carver JR, et al. Task Force 5: Stratification of patients into high-, medium-, and low-risk subgroups for purposes of risk factor management. J Am Coll Cardiol 1996;27:1007-1019.

97. Cohnn PF. Prognosis and treatment of asymptomatic coronary artery disease. J Am Coll Cardiol 1983;1:959-964.

98. Libby P, Theroux P. Pathophysiology of coronary artery disease. Circulation 2005;111:3481-3488.

99. Serruys PW, Kutryk MJB, Ong ATL. Coronary-artery stents. N Engl J Med 2006;354:483-495.

100. Cutlip D, Chauhan M, Baim D, et al. Clinical restenosis after coronary stenting: Perspectives from multicenter clinical trials. J Am Coll Cardiol 2002;40:2082-2089.

101. Popma JJ, Tulli M. Drug-eluting stents. Cardiol Clin 2006;24:217-231.

102. Iakovou I, Schmidt T, Bonizzoni E, et al. Incidence, predictors, and outcome of thrombosis after successful implantation of drug-eluting stents. JAMA 2005;293: 2126-2130.

103. Heublein B, Rohde R, Kaese V, et al. Biocorrosion of magnesium alloys: A new principle in cardiovascular implant technology? Heart 2003;89:651-656.

104. Serruys PW, Ong ATL, Morice M-C, et al. Arterial Revascularisation Therapies Study. II. Sirolimus-eluting stents for the treatment of patients with multivessel de novo coronary artery lesions. Eurointervention 2005;1:147-156.

105. McFadden EP, Stabile E, Regar E, et al. Late thrombosis in drug-eluting coronary stents after discontinuation of antiplatelet therapy. Lancet 2004;364:1519-1521.

106. Prinzmetal M, Kennamer R, Merliss R, et al. A variant form of angina pectoris. Am J Med 1959;27:375-388.

107. Cannon CP, Braunwald E. Unstable angina and non-ST elevation myocardial infarction. In: Libby P, Bonow RO, Mann DL, et al, eds. Braunwald's Heart Disease: A Textbook of Cardiovascular Medicine, 8th ed. Philadelphia, PA: Elsevier; 2008: 1319-1340.

108. Burnsides C, Edwards JC, Lansing AI, et al. Arteriosclerosis in the intramural and extramural portions of coronary arteries in the human heart. Circulation 1956;13: 235-241.

109. Tsujita K, Maehara A, Mintz GS, et al. Comparison of angiographic and intravascular ultrasonic detection of myocardial bridging of the LAD coronary artery. Am J Cardiol 2008;102:1608-1613.

110. Pepine CJ, Balaban RS, Bonow RO, et al. Women's Ischemic Syndrome Evaluation: Current status and future research directions: Report of the National Heart, Lung and Blood Institute workshop: October 2-4, 2002: Section 1: Diagnosis of stable ischemia and ischemic heart disease. Circulation 2004;109:e44-e46.

111. Bugiardini R, Bairey Merz CN. Angina with "normal" coronary arteries: A changing philosophy. JAMA 2005;293:477-484.

112. Kemp HG, Kronmal RA, Vlietstra RE, et al. Seven year survival of patients with normal or near normal coronary ateriograms: A CASS registry study. Am Coll Cardiol 1986;7:479-483.

113. Panting JR, Gatehouse PD, Guang-Zhong Y, et al. Abnormal subendocardial perfusion in cardiac syndrome X detected by cardiovascular MRI. N Engl J Med 2002;346: 1948-1953.

114. Suwaidi JA, Hamasaki S, Higano ST, et al. A. Long-term follow-up of patients with mild coronary artery disease and endothelial dysfunction. Circulation 2000;101: 948-954.