The Missed MI: Understanding the Limits of Biomarkers and CT Angiography
I remember when the Goldman myocardial infarction (MI) algorithm came out, quickly followed by a seven-button hand-held "calculator." It promised to reduce all decision-making regarding ED chest pain patients to seven yes or no questions. But when you looked into the mathematics, if you answered no or negative to all of the questions, it indicated a 4% chance of acute cardiac ischemia. So, what would you do with this information? Could you tell the patient that there was only a 4% chance of a heart attack, so it was OK to go home?
My short summary really is not fair to the Goldman algorithm. It was not designed to determine who among ED chest pain patients could be discharged vs. who needed to stay. So it is with most ancillary tests and aids for chest pain; they assist but do not replace experienced physician judgment. This issue concludes our four-part series with a review of the use of cardiac biomarkers, chest pain units, ACS risk score calculators, and CT angiography. I hope you find this series useful.
J. Stephan Stapczynski, MD, FACEP
Serum Cardiac Markers
Cardiac biomarkers in common use to diagnose acute myocardial infarction (MI) include myoglobin, creatine kinase, creatine kinase-mb isoenzyme (CK-MB), cardiac troponin I, and troponin T. These proteins are found inside muscle tissue and function as enzymes except for troponin, which is part of the contractile apparatus. When a cardiac myocyte dies, the intracellular contents leech out into the bloodstream at rates depending on the protein's size and original location in the cell. Myoglobin is smallest and therefore appears first after only 1-3 hours, with peak levels attained in about 6-7 hours.1 CK-MB levels begin to rise 3-12 hours after cell death but do not peak until about 10-24 hours. (See Figure 1.)
 |
Cardiac Troponin I (CTnI) and T (CTnT) are the current gold standard for diagnosis of acute MI as endorsed in 2000 by the American College of Cardiology and European Society of Cardiology.2 This change in the diagnostic criteria for MI to include enzyme level changes without electrocardiogram (ECG) changes has increased the number of acute MIs diagnosed by about 30%.3 A single troponin level greater than the 99th percentile of normal is the accepted criteria for defining myocardial necrosis.4
Since many physicians often obtain both CK-MB and troponin levels on patients being evaluated for chest pain, several studies have examined the importance of discordant biomarker levels.5 In other words, if the CK-MB is positive but the troponin is negative, or vice versa, is it any less significant than if both were elevated? Large studies have established that an elevated troponin places patients at higher risk regardless of other biomarker levels,6 with a nearly linear correlation with the troponin level and risk for cardiac events/mortality.7 Conversely, Storrow, et al. found in their study that an elevated CK-MB level with a negative troponin also places patients at higher risk for ACS.5 Thus, either elevated troponin levels or CK-MB levels appear to be significant.
Elevated cardiac biomarkers have diagnostic utility, but, like a non-diagnostic ECG, negative biomarker levels do not rule out the presence of coronary disease or unstable angina. Roughly 33% of patients with unstable angina will have elevated troponin levels but normal CK-MB levels; the other 66% have no biomarker elevations.7 Cardiac biomarkers should be interpreted in the context of the patient's symptoms and ECG.
Because of the time it takes for the serum level to reflect the clinical situation, biomarker levels may take 4-6 hours to reflect the damage from the large MI currently underway. (See Figure 1.) Only about 20-30% of acute MI will have elevated biomarkers on the initial assay.8,9
There are at least five different assays for troponin I, and their 99th percentile cutoffs vary from 0.022 µg/L to 0.08 µg/L,9 which is important to remember if working in more than one ED. Biomarkers are most useful as clinical decision aids in patients with intermediate or low pre-test risk.10 Patients who are presenting with a clinical history very suggestive of ACS, numerous risk factors, and new ECG changes are unlikely to be discharged if their troponin levels are negative.
Elevated Troponin Levels: Accurate Interpretation
There are multiple situations in which an elevated troponin may not indicate myocardial ischemia. In other words, cardiac troponin levels are "cardiac-specific but not disease-specific."11 While most cases of mild, non-rising troponin elevations likely are due to a small amount of myocardial necrosis ("troponin leak"), one should correlate the finding of an elevated troponin level with the patient's clinical situation. The American College of Cardiology/European Society of Cardiology definition of MI includes elevated troponin level in the context of clinical conditions suggesting myocardial ischemia (i.e., chest pain) and/or with ECG changes suggesting myocardial ischemia.2 However, the author of this consensus document has pointed out that some physicians have focused on the elevated troponin level alone to diagnose acute MI, which is not consistent with the guidelines.12
Mild troponin elevations are surprisingly common. One study of 883 hospitalized patients older than 18 years of age found that 35% had troponin elevations unrelated to ACS during their stay.13 While these patients' troponin levels were not induced by ACS, 77% later were found to have flow-limiting coronary lesions.13 Thus these non-ACS troponin elevations help to identify patients who are at much greater risk for having coronary disease and likely to be at risk for future cardiac events.13
Yet, not every study finds this result. Another study of sepsis patients with elevated troponin levels found that those with elevations were at increased risk of death but not from cardiac sources.14 Further, it also is unlikely that a patient with mild troponin bumps due to pericarditis would be at risk for coronary disease.
There are four general situations in which troponin levels can be elevated: in acute MI either from primary coronary disease (i.e., classic STEMI) or from ischemia secondary to decreased oxygen supply or increased demand; other diseases of the heart; diseases of other organ systems; or actual false-positive laboratory values.15
Acute MI triggered by secondary ischemia includes conditions such as severe anemia, arrhythmias, prolonged tachycardia (pulmonary embolism, atrial fibrillation, etc.), and hypertensive or hypotensive episodes. There also are reports of non-disease conditions that have been associated with troponin elevations, including vigorous exercise. One study of runners in the Boston Marathon found that 68% of runners had some degree of post-race troponin elevation,16 with 11% of the elevations that were diagnostic of acute MI.16 Similar changes also have been noted in elite cyclists and triathletes.16
Other heart diseases, such as congestive heart failure (CHF), myocarditis, and pericarditis, are known to cause troponin elevations. In CHF patients, the degree of elevation does appear to correlate with clinical severity of the disease.17 Cardiac trauma, either iatrogenic as in surgery or from blunt chest trauma, also is associated with troponin elevations. Troponin elevation is not a consistently useful prognostic indicator. Some studies show correlation with increased risk of arrhythmia and ventricular dysfunction,18 while others show no correlation with complications or outcome.19
Troponin elevations also have been documented resulting from many other systemic illnesses, including conditions of sympathetic hyperactivity (ischemic stroke, intracranial hemorrhage, subarachnoid hemorrhage, Takotsubo syndrome), end-stage renal disease (ESRD), and from chemotherapy agents, carbon monoxide poisoning, and certain biologic toxins (i.e., jellyfish, scorpion, centipede).11
Troponin levels (and CK-MB levels) in renal failure patients on hemodialysis deserve special attention, as ESRD patients commonly are seen in the ED, and troponin elevations are reported in a high proportion of otherwise asymptomatic patients.20 Elevated CTnT levels were seen in 42%, and elevated CTnI levels were seen in 15%20 of ESRD patients.
Elevation of either troponin subunit in ESRD patients correlates strongly with increased mortality. One study found ESRD patients with elevated troponin levels had 1-year cardiac and all-cause mortalities of 26% and 52%, respectively.21 The basis of the elevated troponin levels is thought to be due to increased coronary disease and subclinical MI, mild global ischemia, or to another as yet unknown process.22
Uncommonly, true false-positive elevations are seen at times due to the presence of substances in the serum that interfere with the troponin immunoassay. Examples include heterophil antibodies, autoantibodies, rheumatoid factor, or high levels of free hemoglobin/bilirubin, lipid levels, and alkaline phosphatase.11 Troponin levels falsely elevated by as much as 20% also have been described from use of dilution solutions not recommended by the assay manufacturer.23
Careful Use of Serum Markers: Is One Set OK?
Although clinical policies/guidelines from multiple medical organizations, including the American College of Cardiology (ACC), American Heart Association (AHA), and the American College of Emergency Physicians (ACEP), are to check serial sets of cardiac biomarkers to accurately diagnose acute MI, it was noted as far back as 2001 that many physicians across the United States sometimes obtain a single troponin prior to discharge of ED chest pain patients.24
Clinical Guidelines from ACC/ACEP. While these guidelines are not mandatory rules to govern one's practice, one can be assured that plaintiff lawyers are well aware of these guidelines and will use them to their advantage whenever possible. Given that missed MIs represent the single most costly suits in terms of dollar payouts,25 and these guidelines are derived from extensive research, one would be cavalier to ignore them.
The AHA definition of ischemic heart disease published in 2003 defines an "adequate set of biomarkers" as "at least 2 measurements of the same marker taken at least 6 hours apart."26 Gibler, et al published in 2005 a reiteration of the ACC/AHA guidelines that "serial testing in the ED, at 3 and 6 hours, and at an interval of 6 to 10 hours in-hospital, is necessary to exclude myocardial injury."27
The most recent guidelines from the ACC/AHA were published in 2007,4 and suggest an initially negative troponin should be followed by a second level 8-12 hours later. The only Level A recommendation from the current ACEP clinical policy on management of ACS/nonSTEMI is to "not utilize cardiac serum markers to exclude non-AMI acute coronary syndromes (i.e., unstable angina)."28 The ACEP guidelines also state that when the time of symptom onset is unclear or unknown, then the time of ED presentation should be used as time of symptom onset.28
Published Reviews. Other cardiology-based reviews have appeared in well-known journals (New England Journal of Medicine, Journal of the American Medical Association), giving clear statements on the use of cardiac biomarkers. In 1997, Hamm et al., in a paper on the use of at least two troponin levels for triage of ED chest pain patients, noted that "a single test at the time of arrival is inadequate for clinical decision making."29 Antman et al published in 1995 that "single (troponin) values should not be used to rule out infarction."30 Lastly, in 2000 Lee et al. stated that "single values should not be used to rule out myocardial infarction," even referring to patients whose symptoms had begun more than 12 hours prior to presentation.31
 |
The reasons for these statements are based on the inherent time delay for detectable troponin (or CK-MB) levels to appear in the serum after myocyte death. (See Figure 1.) Initial biomarker levels taken on patient arrival have a reported range of 10% to 78% for sensitivity of detecting acute MI.32 This likely results from great variability between when the patient's symptoms begin vs. when actual myocyte death occurs.
When Can a Single Troponin Help? These previous statements beg the question, is there any circumstance in which single negative troponin levels can be useful? A clear example is in evaluation of a chest pain episode that occurred several days before presentation.32 Because troponin levels remain elevated for 5-7 days after acute MI, these kinetics would allow one to say with reasonable certainty that a three-hour chest pain episode two days prior did not result in myocardial infarction. The ACEP guidelines give a Level B recommendation on single troponin levels.28 They recommend use of a single negative CK-MB or troponin drawn 8-12 hours after symptom onset, using time of ED presentation when symptom onset is unavailable/unclear.
Checking a Second Set Before 6 Hours. Several groups have studied the feasibility of not waiting the full six hours between sets of biomarkers known as delta measurements. In other words, the biomarkers are repeated at more frequent intervals to speed up decision-making and ED efficiency. Some have used 2-hour changes in CK-MB levels alone,34 while others have used longer time intervals of 3 hours.35 Yet others have used combinations of delta myoglobin and troponin levels, done with 90-minute intervals.36,37 Nearly all of these studies have shown encouraging results, with sensitivities of 92-99%, even at shorter intervals.
However, all of the studies have used different criteria for inclusion of patients and, more importantly, different formulas for defining what is a "positive" delta level. These studies define the amount of change required to become significant, even though in most cases the second biomarker level may still be in the "normal" range.34,35,37 For example, Fesmire used a delta of CK-MB of only 1.6 ng/mL or greater.34 These approaches may be helpful, but clear criteria for positive delta levels, defined time intervals, and which biomarkers to use will require further consensus.
Another approach that can be applied to current practice is taken by MacRae et al.38 They simply took more frequent serum samples before the 6-hour mark and examined how many biomarker sets became elevated using the AHA 99th percentile definition of elevated biomarkers.38 They concluded that using the 3-hour interval was a useful alternative to waiting 6 hours. Further, they found that a 1-hour interval allowed diagnosis of 80% of acute MIs as long as at least one set was taken > 6 hours after symptom onset.38 Currently, many chest pain centers employ a biomarker rule-out protocol of 3 sets over a 9-hour period.39
Newer Serum Markers
Other proteins evaluated for usefulness as serum markers in evaluation of chest pain/acute MI patients include: vascular cell adhesion molecule, P-selectin, interleukin-1, tumor necrosis factor-a, serum amyloid A, ischemia-modified albumin, B-type natiuretic peptide (BNP), and pregnancy-associated plasma protein A.40 Many of these markers are being investigated as a result of work that suggests acute plaque rupture may be related more to inflammation of the plaque rather than the degree of luminal stenosis.41 Thus, atherosclerosis is being thought of more as an inflammatory process rather than a structural disease.
High-sensitivity C-reactive protein (hsCRP) is the best known of these markers. The concept is that when hsCRP levels are elevated, the patient is experiencing an increased inflammatory state and therefore is at higher risk for an acute event. Currently, hsCRP levels have been shown to be independent predictors of future cardiac events, including acute MI, CHF, and cardiac death. One study found that men with the highest hsCRP levels ( > 2.1 mg/L) were 3 times as likely to develop MI as controls even when matched for other traditional risk factors.42 Unfortunately, studies on the utility of hsCRP in the acute setting have failed to show any benefit.43 Thus, hsCRP levels can be used to predict one's risk of future events and may be used in the future to guide risk factor modification, but currently this marker does not have a role in diagnosis or therapy for acute events.1
 |
Chest Pain Units (CPU)
Chest pain centers arose as an answer to the need for full admission of large numbers of patients who present to the ED with chest pain without obvious cause but are not deemed entirely safe for discharge. Previously, these patients were admitted for inpatient work-ups at a nationwide cost calculated in 1984 of $5-8 billion, but, at most, only 20-30% ultimately would be diagnosed with ACS.44 Typically, a patient now can be admitted to observation status in a CPU, have a full set of three biomarkers obtained, and in many cases cardiology consultation with optional stress testing all in 12-24 hours.
Exercise testing of the patient before discharge is preferable but not mandatory. When patients are discharged without a stress test, however, it is recommended that one be performed soon on an outpatient basis (i.e., 48 hours).45 It also is possible in lower-risk patients that no stress test may need to be done, as evidenced in a recent study that found no adverse events in 175 such CPU patients.46
Another approach is immediate exercise testing of patients without rule-out by serum markers.39 A study of 1000 patients using this approach found that 13% of the stress tests were positive, 64% were negative, and 23% were non-diagnostic.39 No adverse events were noted during the exercise testing or in the following 30 days afterward. This represents another possible approach, but one must be careful in screening patients into this pathway to ensure they truly are low-risk.47
The average cost of observation vs. inpatient management of chest pain patients is estimated to be 45-80% less.48,49 It was estimated in 1997 that roughly 30% of U.S. EDs use a chest pain unit.50 The chest pain center concept likely represents the most realistic approach to reduce the number of missed MIs discharged from the ED.
Risk Score Calculators
The first well-known risk calculator was the Goldman protocol, developed in the early 1980s.51 Since then, multiple other protocols have been proposed. Risk score calculators have been developed as decision aids to help identify patients at risk for acute MI. The Goldman protocol was developed only to predict likelihood of increased risk in patients who were suspected to have an acute MI and had a sensitivity around 90%. This protocol does seem to be useful in deciding which level of care (CCU, monitor bed, or observation) patients may require while admitted to the hospital.52 However, its main flaw is that it was not made to detect patients with unstable angina.53 The Goldman protocol later was modified to be used for recognition of ACS patients,54 but this modified protocol still focuses primarily on only the ECG findings and clinical condition at time of presentation. An attempt to add a single troponin level combined with the Goldman protocol also was found not to be sensitive enough for use in the ED.24 A recent review of risk score models concluded the modified Goldman protocol has less than a 90% sensitivity for ACS patients.55
Pozen developed a decision aid using seven clinical and ECG variables, which helped reduce the admission rate to the CCU but did not lower the inappropriate discharge rate of 3%.56 Pozen's equation was modified by Selker in 1998 to develop a risk score named ACI-TIPI (Acute Cardiac Ischemia Time-Insensitive Predictive Instrument).57 This calculation can be added into standard ECG machines so that the score is printed on the ECG header. This protocol appears to increase the safe discharge rate of unsupervised residents but did not change the rate for attending physicians.58 Thus, the utility of this tool appears limited to less experienced physicians. Further, the ACI-TIPI is designed to have a 10% chance of missed ACS or MI, which currently is much higher than the 2-4% rate observed in the United States.59
The thrombolysis in myocardial infarction (TIMI) risk score60 and Sanchis scores61 also have been used in ED chest pain patients, although both originally were designed on the assumption that patients were likely to have ACS and to aid in prediction of short-term prognosis (TIMI) or 1-year adverse events.62 Both the TIMI score and the Sanchis score recently were studied in acute chest pain patients with non-diagnostic ECGs and initial negative biomarkers and were found to be less than 90% sensitive.55,62
The Vancouver Chest Pain Rule published in 2006 is designed to identify patients with ACS and allow safe discharge of 30% of patients presenting with chest pain in only 2-3 hours.58 This protocol is based partly on a single CK-MB measurement (albeit set at a lower threshold) at presentation to allow discharge of some patients, or on a 2-hour delta level of the CK-MB or troponin. This protocol is designed to miss fewer than 2% of these patients with ACS, which is similar to current estimates of the rate of missed ACS in the United States.59
The bottom line on risk score calculators is that despite initial enthusiasm,63 there currently are no algorithms that have proven useful in general ED populations to replace experienced physician judgment.
Cardiac CT Scans
The most promising development for rapid determination of chest pain patients who can be discharged safely from the ED is that of the coronary CT scan. The 64-slice multidetector-row CT (MDCT) systems allow for both high-speed and high-resolution CT imaging. The CT scan is gated with the ECG signal so that images taken while the heart is in motion are subtracted out, thus providing clear images of the coronary vessels. A slow and regular heart rate is required for the study, and most patients are given beta-blockers to reach the required rate of < 65 bpm. Nitroglycerin also is given to maximally dilate coronary arteries. Newer generation dual-source CT scanners recently have been studied for CT coronary angiography in patients with atrial fibrillation or sinus arrhythmias but achieve diagnostic accuracies of only 88-90%.64
Radiation Exposure Risks
An important aspect of cardiac CT scanning is the radiation doses involved. While awareness of the possible increased lifetime risk of cancer from radiation with diagnostic procedures has increased in the last few years, many physicians and most patients are unaware of the potential risk of radiation exposure with CT scans.65 The roughly 62 million CT scans obtained in the United States in 2006 were only 15% of the radiographic procedures performed, but were responsible for > 50% of the radiation exposure.66
Patient age is the most important factor that affects the risk of developing cancer from even a single CT scan. In general, the younger the patient, the higher the potential risk, because children's actively dividing cells are more radiosensitive and children have more years to live in which cancer may arise. Full-scale epidemiologic studies to accurately delineate risk for all patient populations are not available,67 and current data on possible cancer risk are derived from atomic bomb survivors.67 They estimate that patients younger than one year of age have a lifetime risk of cancer death (primarily brain cancer) of 0.07% from a single head CT and 0.14% from a single abdominal CT.67 Risks from head CT drop dramatically with age and are just over 0.01% at age 15 years, while risks for abdominal CT do not drop under 0.02% until age 35 years. It is estimated that a full-body scan, such as the pan-scan in a trauma patient, gives a 45-year-old patient a 1 in 1,250 risk of dying from cancer.68
Single head CT scans yield radiation doses of about 5-20 mSv, and single abdominal CT doses range from 10-20 mSv. However, exposure doses can be several times higher for pediatric head CT where smaller bodies/organs can be exposed to the same amount of radiation as adults. For reference, the yearly federal limit for workers in the United States is 50 mSv and is only 20 mSv in the UK. A 20 mSv dose is approximately 130 times the dose for a lateral chest film, 2000 times that of a PA chest, and 4000 times the dose one's brain is exposed to from a single dental radiography.67 Another point of reference is that a 20 mSv dose is equivalent to roughly 6 years of normal background radiation exposure.
Coronary CT imaging with a 16-slice scanner compared to a 64-slice machine had radiation doses of 5-10 mSv and 5-15 mSv, respectively.69 While these doses are not much higher than other types of CT scans, a recent paper found surprising variability in the dose measured at 50 different study sites; the average radiation dose was 12 mSv but ranged from as little as 5 mSv to as high as 30 mSv.70 A study of exposure in the "triple rule-out" CT used to evaluate a patient's aorta, pulmonary vessels, and coronary vessels simultaneously found an estimated average dose of 18 mSv (range 10-31 mSv).71 Obese patients received nearly 25% more radiation exposure in that study (21 mSv vs. 15 mSv).72 These numbers compare to the average radiation dose from a nuclear stress perfusion scan of 15-25 mSv and 2-6 mSv for invasive angiography.73
Calcium Scores
Coronary imaging by CT scans is done using two different approaches: detection of coronary calcifications only and, most recently, direct visualization of the coronary lumen. The older technique detects the presence of calcium deposits in the coronary arteries, is done without any contrast dye, and there is no visualization of the coronary lumen. The principle for this test is that coronary calcification is almost exclusively from atherosclerosis, with the only real exception being patients with renal failure. A "calcium score" is calculated, which is thought to be a good overall indicator of total atherosclerotic burden, with higher scores equaling more calcium deposits. The commonly used Agatston Score ranges are: 0 equals no evidence of coronary disease, 1-10 equals minimal disease, 11-100 equals mild disease, 101-400 equals moderate disease, and more than 400 equals extensive disease.
The absence of coronary calcium excludes the presence of significant stenotic lesions with high predictive value.74 Low calcium scores also equate with very low risk for coronary heart disease regardless of other classic risk factors.75 Likewise, several large studies have found that higher calcium scores do predict increased risk of MI or death in the following 3-5 years, even in asymptomatic people.75 Unfortunately, high calcium scores do not correlate well with the presence of hemodynamically significant stenosis at the time of the scan. Thus, the sensitivity of coronary calcification to predict coronary disease is about 90%, but since calcifications can make image interpretation difficult, the sensitivity for diagnosing obstructive disease is only about 50%.76
The most recent recommendations from the ACC/AHA consensus statement is for calcium scoring to be used only in asymptomatic patients with intermediate risk for coronary heart disease and is not recommended for asymptomatic patients with low or high risk.77
CT Coronary Angiography
With 64-slice multidetector-row CT (MDCT) systems, the resolution of coronary CT scans has reached the point where coronary artery lumens can be assessed. In some cases, these CT images can be used in place of traditional invasive coronary angiography. Coronary angiography using 64-slice CT machines has sensitivities and specificities ranging from 86-99% and 93-97%, respectively, and anywhere from 2-12% of coronary segments nondiagnostic.76 With current technology, only coronary arteries greater than 1.5 mm are well-visualized and, therefore, disease in vessels less than 1.5 mm may not be evaluated.78
CT angiography is accurate in ruling out disease with negative predictive values of 93-100% and also can evaluate extralumenal lesions that have previously only been seen with intravascular ultrasound. However, it is less accurate in defining existing disease, with positive predictive values of anywhere from 50-100%.76 CT angiography suffers from a relatively high number of false-positives (up to 30%) and thus tends to overestimate disease severity when compared to invasive angiography. Other problems with accuracy are related to artifacts from reduced resolution caused by pronounced coronary calcifications and arrhythmias.
CT Evaluation of Chest Pain Patients
The only randomized study to date comparing standard ED treatment to CT angiography evaluated 99 patients with CT and 98 with standard care.79 The study found that CT identified or excluded coronary disease as the cause of chest pain in 75% of the patients; 67 had normal coronary arteries, and 8 had severe disease.79 The other 25% had either intermediate coronary disease or nondiagnostic CT studies and underwent further evaluation by stress testing. An editorial comment of this study expressed concern that the CT group underwent four times as many major invasive procedures, including catheterization, stenting, or bypass surgery, as the control group but "gained no identifiable medical benefit either in the short term or at 6-month follow-up."80 The concern for over-treatment in the absence of medical benefit is a legitimate one, but it is not clear how a patient who required bypass surgery did not benefit medically from being identified by CT angiography.
In another recent study of 568 ED chest pain patients, 285 had coronary CT immediately in the ED while another 283 had coronary CT after observation.81 The study was not randomized, as patients were admitted to the observation arm based on the availability of coronary CT (nights/weekends). Fully 75% of the group with immediate CT were discharged afterward with no adverse events noted in the following 30 days.81 The authors suggest that CT angiography may be used to shorten the time it takes to determine if a patient is safe for discharge after presenting to the ED with chest pain. It would help ED physicians to identify a subgroup of patients who can be discharged safely from the ED even without serial enzymes or a stay in an observation unit. This idea may prove to be true, but a problem with this study is that there were no coronary events in either group. Thus no MIs were missed, but none were found either.
The "Triple Rule Out" Scan. CT angiography can be used to simultaneously evaluate the coronary anatomy and rule out both aortic disease and pulmonary embolism. These studies have been referred to as the "triple rule out."82 A recent study of the triple rule out CT on 197 ED chest pain patients who had nondiagnostic ECGs and negative initial biomarkers found 65% had no coronary disease, 23% had mild (< 50%) stenosis, and 11% had moderate to severe disease.82 However, there was one case of severe coronary disease that was "missed" on the CT images. In addition, 10% of patients had "suboptimal" imaging of at least one coronary vessel, and there were two positive follow-up stress tests in the group of patients with no disease on CT scan.82 CT imaging did identify 11% of patients with other diagnoses than ACS that explained their symptoms, such as pneumonia, pulmonary embolism, hiatal hernia, aortic dissection, metastatic pulmonary disease, myocarditis, and pancreatitis. Another 14% had clinically important non-coronary findings such as aortic aneurysm, emphysema, cardiomyopathy, adrenal mass, pulmonary mass, and pulmonary hypertension.82
The Bottom Line for CT Angiography
Coronary CT angiography undoubtedly will be helpful in rapidly identifying a subset of ED chest pain patients with no coronary artery disease. It ultimately may be that identification of patients without coronary artery disease or discovering disease where it was not suspected may be the strong suit for coronary CT. One recent study suggested we may be underestimating the extent of coronary disease in populations considered to be low risk based on classic Framingham risk factors. This study found a disturbingly high rate of coronary plaques/obstructive disease in patients classed as low risk (44/16%) and intermediate risk (75/34%) by classic criteria.83 However, like all other tests presently available, there still are imperfections. To start, most CT studies so far only evaluate non-obese patients with body mass indexes (BMIs) in the mid-twenties. One of the only studies to report on BMI of their patients found that those with BMI of 30 kg/m2 or more accounted for all but one of their inaccurate results.84
Obtaining a negative CT result will not mean that a patient can be discharged safely. For example, even in the best studies there are segments of the coronary arteries that cannot be evaluated on the CT scan due to technologic limits.78 This number has been as high as 13% of vessels imaged in a recent study, and motion artifacts and/or calcium deposits are the two most common contributing sources.85 Coronary calcifications typically are a result of coronary disease, but motion artifacts may not allow one to assess patient risk. More importantly, vessels 1.5-2.0 mm are not always evaluated by CT angiography. Thus, the question of how clinically important stenoses in these smaller vessels turn out to be needs to be answered. Further, patients with coronary spasm who have normal vessel anatomy likely will not be diagnosed by CT scan. A recent study of CT angiography in cocaine chest pain patients confirmed that the patients in the study did not have statistically more coronary disease than controls.86 However, these patients still were at a six-fold higher risk of ACS, presumably due to coronary spasm from their cocaine use. Patients with subendocardial ischemia caused by a hypertrophied left ventricle are another example of patients at risk for MI with normal vessel anatomy.
Another and perhaps more important aspect is that even if CT coronary angiography becomes 100% accurate to define coronary anatomy, this is only part of the question. For patients in whom coronary disease is found, the second half of that question becomes, is the coronary lesion seen physiologically important? In other words, is this lesion causing myocardial ischemia to explain the patient's symptoms?
Hollander's recent study of low-risk patients with CT angiography found that patients with less than 50% lesions on CT angiography had normal stress tests.81 Likewise, those with greater than 80% stenosis all had abnormal stress tests. The difficult patients were those with 50-70% lesions in whom some positive and some negative stress tests were obtained. Thus, even though their anatomy was defined, it was not clear based on anatomy alone who needed further intervention.
This principle is replicated by a recent study of 100 patients with coronary disease that found patients who were diagnosed with obstructive disease on CT had a 63% cardiac event rate (MI, revascularization) in the next 16 months.87 However, patients diagnosed with non-obstructive disease still had an 8% cardiac event rate in the same time period. This 8% number is much higher than the 0.4% risk of MI/death per year seen with negative nuclear perfusion scans.88 One likely would need to combine some kind of functional study to select out patients with higher risk for intervention.
Current Recommendations in the Literature
Recently a group of eight different cardiology and radiology societies collectively published consensus guidelines on indications for CT angiography.89 Study indications were described as appropriate, inappropriate or uncertain. The relevant opinions for the ED include only two appropriate indications: for patients with intermediate pretest probability of coronary disease and no ECG changes or elevated biomarkers, or of patients with uninterpretable ECG/inability to exercise.89 CT study of high or low pretest probability patients was deemed uncertain, as was the triple rule out study. Only patients with diagnostic ECG changes or biomarker levels were deemed inappropriate for CT evaluation.89
Future Directions
On the horizon, combining coronary CT imaging with perfusion imaging, as with single photon emission CT (SPECT) or positron emission tomography (PET) scans, may replace current perfusion scans with sestamibi or thalium-201.90 This may allow a single study that would detect coronary lesions and simultaneously determine their hemodynamic significance in myocardial perfusion. Cost factors, radiation exposure, and limited availability of PET scans are all issues to overcome before these studies can be widely used. Another possible advancement is that of dual- energy CT that allows assessment of organ perfusion as well as structure.90
References
1. Aviles JM, Aviles RJ. Advances in cardiac biomarkers. Emerg Med Clin North Am 2005;959-975.
2. Alpert JS, Thygesen K, Antman E, et al. Myocardial infarction redefined: A consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. J Am Coll Cardiol 2000;36:959-969.
3. Bhatheja R, Mukherjee D. Acute coronary syndromes: Unstable angina/non-ST elevation myocardial infarction. Crit Care Clin 2007;23:709-735.
4. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007 Guidelines for the Management of Patients With Unstable Angina/Non-ST-Elevation Myocardial Infarction A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/Non-ST-Elevation Myocardial Infarction). Circulation 2007;116:e148-e304.
5. Storrow AB, Lindsell CJ, Han JH, et al. Discordant cardiac biomarkers: frequent and outcomes in emergency department patients with chest pain. Ann Emerg Med 2006;48:660-665.
6. Newby LK, Roe MT, Chen AY, et al. Frequency and clinical implications of discordant creatine kinase-MB and troponin measurements in acute coronary syndromes. J Am Coll Cardiol 2006;47: 312-318.
7. Antman EM, Tanasijevic MJ, Thompson B, et al. Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med 1996;335:1342-1349.
8. Ghaemmaghami CA, Brady WJ. Pitfalls in the emergency department diagnosis of acute myocardial infarction. Emerg Med Clin North Am 2001;9:351-369.
9. Melanson SEF, Tanasijevic MJ, Jarolim P. Cardiac troponin assays. A view from the clinical chemistry laboratory. Circulation 2007;116:e501-e504.
10. Saenger AK, Jaffe AS. The use of biomarkers for the evaluation and treatment of patients with acute coronary syndromes. Med Clin North Am 2007;91:657-681.
11. Tsai SH, Chu SJ, Hsu CW, et al. Use and interpretation of cardiac troponins in the ED. Am J Emerg Med 2008;26:331-341.
12. Alpert JS. Lab tests don't make diagnoses, doctors do. Am J Med 2008;121:87-88.
13. Blich M, Sebbag A, Attias J, et al. Cardiac troponin I elevation in hospitalized patients without acute coronary syndromes. Am J Cardiol 2008;101:1384-1388.
14. Ammann P, Maggiorini M, Bertel O, et al. Troponin as a risk factor for mortality in critically ill patients without acute coronary syndromes. J Am Coll Cardiol 2003;41: 2004-2009.
15. Goran KP. Simple and practical classification of elevated troponin values. Am J Emerg Med 2008;26:951-952.
16. Fortescue EB, Shin AY, Greenes DS, et al. Cardiac troponin increases among runners in the Boston Marathon. Ann Emerg Med 2007;49:137-143.
17. Yilmaz A, Yalta K, Turgut OO, et al. Clinical importance of elevated CK-MB and troponin I levels in congestive heart failure. Adv Ther 2006;23:1060-1067.
18. Rajan GP, Zellweger R. Cardiac troponin I as a predictor of arrhythmia and ventricular dysfunction in trauma patients with myocardial contusion. J Trauma 2004; 57:801-808.
19. Bertinchant JP, Polge A, Mohty D, et al. Evaluation of incidence, clinical significance, and prognostic value of circulating cardiac troponin I and T elevation in hemodynamically stable patients with suspected myocardial contusion after blunt chest trauma. J Trauma 2000;48:924-931.
20. Choy JB, Armstrong PW, Ulan RA, et al. Co cardiac troponins provide prognostic insight in hemodialysis patients? Can J Cardiol 2003;19:907-911.
21. Kontos MC, Garg R, Anderson FP, et al. Outcomes in patients admitted for chest pain with renal failure and troponin I elevations. Am Heart J 2005;150:674-680.
22. deFilippi CR, Thorn EM, Aggarwal M, et al. Frequency and cause of cardiac troponin T elevation in chronic hemodialysis patients from study of cardiovascular magnetic resonance. Am J Cardiol 2007;100:885-889.
23. Tze-Kiong E, Jong YJ, Chiang CH, et al. Falsely elevated cardiac troponin I attributed to inappropriate dilution material. Am J Emerg Med 2008;27:123-125.
24. Limkakeng A Jr, Gibler WB, Pollack C, et al. Combination of Goldman risk and initial cardiac troponin I for emergency department chest pain patient risk stratification. Acad Emerg Med 2001;8:696-702.
25. Katz DA, Williams GC, Brown RL, et al. Emergency physician's fear of malpractice in evaluation of patients with possible acute cardiac ischemia. Ann Emerg Med 2005:46: 525-533.
26. Leupker RV, Apple FS, Christenson RH, et al. Case Definitions for Acute Coronary Heart Disease in Epidemiology and Clinical Research Studies A Statement From the AHA Council on Epidemiology and Prevention; AHA Statistics Committee; World Heart Federation Council on Epidemiology and Prevention; the European Society of Cardiology Working Group on Epidemiology and Prevention; Centers for Disease Control and Prevention; and the National Heart, Lung, and Blood Institute. Circulation 2003;108:2543-2549.
27. Gibler WB, Cannon CP, Blomkains AL, et al. Practical implementation of the Guidelines for unstable angina/non-ST-segment elevation myocardial infarction in the emergency department. Ann Emerg Med 2005;46:185-197.
28. Fesmire FM, Decker WW, Diercks DB, et al. Clinical policy: Critical issues in the evaluation and management of adult patients with non-ST-segment elevation acute coronary syndromes. Ann Emerg Med 2006;48: 270-301.
29. Hamm CW, Goldmann BU, Heeschen C, et al. Emergency room triage of patients with acute chest pain by means of rapid testing for cardiac troponin T or troponin I. N Engl J Med 1997;337:1648-1653.
30. Antman EM, Grudzien C, Sacks DB. Evaluation of a rapid bedside assay for detection of serum cardiac troponin T. JAMA 1995;273:1279-1282.
31. Lee TH, Goldman L. Evaluation of the patient with acute chest pain. N Engl J Med 2000;342:1187-1195.
32. Brady WJ, Harrigan RA, Chan T. Acute coronary syndromes. In: Marx, JA ed. Rosen's Emergency Medicine: Concepts and Clinical Practice, 6th ed. St. Louis: Mosby; 2006: 1154-1195.
33. Kohn MA, Kwan E, Gupta M, et al. Prevalence of acute myocardial infarction and other serious diagnoses in patients presenting to an urban emergency department with chest pain. J Emerg Med 2005;29: 383-390.
34. Fesmire FM, Percy RF, Bardoner JB, et al. Serial creatinine kinase (CK) MB testing during the emergency department evaluation of chest pain: utility of a 2-hour delta CK-MB of + 1.6 ng/ml. Am Heart J 1998;136:237-244.
35. Young GP, Gibler WB, Hedges JR, et al. Serial creatine kinase-MB results are a sensitive indicator of acute myocardial infarction in chest pain patients with nondiagnostic electrocardiograms: the second Emergency Medicine Cardiac Research Group Study. Acad Emerg Med 1997;4:869-877.
36. Ng SM, Krishnaswamy P, Morissey R, et al. Ninety minute accelerated critical pathway for chest pain evaluation. Am J Cardiol 2001;88:611-617.
37. McCord J, Nowak RM, McCullough PA, et al. Ninety-minute exclusion of acute myocardial infarction by use of quantitative point-of-care testing of myoglobin and troponin I. Circulation 2001;104:1483-1488.
38. Macrae AR, Kavsak PA, Lustig V, et al. Assessing the requirement for the 6-hour interval between specimens in the AHA classification of myocardial infarction. Clinical Chemistry 2006;52:812-818.
39. Amsterdam EA, Leewis WR, Kirk JD, et al. Acute ischemic syndromes: chest pain center concept. Cardiol Clin 2002;20:117-136.
40. Wang TY, AlJaroudi WA, Newby LK. Markers of cardiac ischemia and inflammation. Card Clin 2005;23:491-501.
41. Libby P. Molecular bases of the acute coronary syndromes. Circulation 1995;91: 2844-2850.
42. Ridker P, Rifai N, Pfeffer M, et al. Inflammation, pravastatin, and the risk of coronary events after myocardial infarction in patients with average cholesterol levels. Circulation 1998;98:839-844.
43. Abrams J. C-reactive protein, inflammation, and coronary risk: an update. Cardiol Clin 2003;21:327-331.
44. Pozen MW, D'Agostino RB, Selker HP, et al. A predictive instrument to improve coronary-care-unit admission practices in acute ischemic heart disease: A prospective multicenter clinical trial. N Engl J Med 1984;310:1273-1278.
45. Amsterdam EA, Kirk JD, Diercks DB, et al. Exercise testing in chest pain units: rationale, implementation, and results. Cardiol Clin 2005;23:503-516.
46. Diercks DB, Kirk JD, Turnipseed SD, et al. Assessing the need for functional diagnostic testing in low-risk women with chest pain. Crit Pathw Cardiol 2006;5:64-68.
47. Blomkalns AL, Gibler WB. Chest pain unit concept: rationale and diagnostic strategies. Cardiol Clin 2005;23:411-421.
48. Gomez MA, Anderson JL, Karagounis LA, et al. An emergency department-based protocol for rapidly ruling out myocardial ischemia reduces hospital time and expense: Results of a randomized study (ROMIO). J Am Coll Cardiol 1996;28:25-33.
49. Gaspoz JM, Lee TH, Weinstein MC, et al. Cost-effectiveness of a new short-stay unit to "rule out' acute myocardial infarction in low risk patients. J Am Coll Cardiol 1994; 24:1249-1259.
50. Hoekstra JW, Gibler WB: Chest pain evaluation units: An idea whose time has come. JAMA 1997;278:1701-1702.
51. 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.
52. Reilly BM, Evans AT, Schaider JJ, et al. Impact of a clinical decision rule on hospital triage of patients with suspected acute cardiac ischemia in the emergency department. JAMA 2002;288:342-350.
53. Pope JH, Selker HP. Acute coronary syndromes in the ED: Diagnostic characteristics, test, and challenges. Cardiol Clin 2005;23:423-451.
54. Goldman L, Cook E, Johnson P, et al. Prediction of the need for intensive care in patients who come to the emergency department with acute chest pain. N Engl J Med 1996;334:1498-1504.
55. Manini AF, Dannemann N, Brown DF, et al. Limitations of risk score models in patients with acute chest pain. Am J Emerg Med 2009;27:43-48.
56. Pozen M, D'Agostino R, Selker H, et al. A predictive instrument to improve coronary care unit admission practices in acute ischemic heart disease: A prospective multicenter clinical trial. N Engl J Med 1984; 310:1273-1278.
57. Selker HP, Beshansky JR, Griffith JL, et al. Use of the acute cardiac ischemia time-insensitive predictive instrument (ACI-TIPI) to assist with triage of patients with chest pain or other symptoms suggestive of acute cardiac ischemia: A multicenter, controlled clinical trial. Ann Intern Med 1998; 129:845-855.
58. Christenson J, Innes G, McKnight D, et al. A clinical prediction rule for early discharge of patients with chest pain. Ann Emerg Med 2006;47:1-10.
59. 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.
60. Antman EM, Cohen M, Bernink PJ, et al. The TIMI risk score for unstable angina/non-ST elevation MI: A method for prognostication and therapeutic decision making. JAMA 2000;284:835-842.
61. Sanchis J, Bodi V, Nunez J, et al. New risk score for patients with acute chest pain, non-ST-segment deviation, and normal troponin concentrations. J Am Coll Cardiol 2005;46:443-449.
62. Chase M, Robey JL, Zogby KE, et al. Prospective validation of the Thrombolysis in Myocardial Infarction Risk Score in the emergency department chest pain population. Ann Emerg Med 2006;48:25-259.
63. Lee TH. Chest pain algorithms in the troponin era: Tme to integrate expertise. Am Heart J 1999;138:606-608.
64. Brodoefel H, Burgstahler C, Tsiflikas I, et al. Dual-source CT: Effect of heart rate, heart rate variability, and calcification on image quality and diagnostic accuracy. Radiology 2008;247:346-355.
65. Lee CI, Haims AH, Monico EP, et al. Diagnostic CT scans: assessment of patient, physician, and radiologist awareness of radiation dose and possible risks. Radiology 2004;231:393-398.
66. Mettler FA Jr, Wiest PW, Locken JA, et al. CT scanning: Patterns of use and dose. J Radiol Prot 2000;20:353-359.
67. Brenner DJ, Hall EJ. Computed tomography an increasing source of radiation exposure. N Engl J Med 2007;357: 2777-2784.
68. Brenner D, Elliston C. Estimated radiation risks potentially associated with full-body CT screening. Radiology 2004;232:735-738.
69. Hausleiter J, Meyer J, Hadamitzky M, et al. Radiation dose estimates from cardiac multislice computed tomography in daily practice. Circulation 2006;113:1305-1310.
70. Hausleiter J, Meyer T, Hermann F, et al. Estimated radiation dose associated with cardiac CT angiography. JAMA 2009; 301:500-507.
71. Ketelsen D, Thomas C, Werner M, et al. Dual-source computed tomography: Estimation of radiation exposure of ECG-gated and ECG-triggered coronary angiography. Eur J Radiol 2008;18:592-599.
72. Takakuwa KM, Halpern EJ, Gingold EL, et al. Radiation dose in a "triple rule-out" coronary CT angiography protocol of emergency department patients using 64-MDCT: The impact of ECG-based tube current modulation on age, sex, and body mass index. AJR Am J Roentgenol 2009; 192:866-872.
73. Ibanez B, Badimon JJ, Garcia MJ. Diagnosis of atherosclerosis by imaging. Am J Med 2009;122:S15-S25.
74. Slovis TL, Berdon WE. Panel discussion. Pediatr Radiol 2002;32:242-244.
75. Church TS, Levine BD, McGuire DK, et al. Coronary artery calcium score, risk factors, and incident coronary heart disease events. Atherosclerosis 2007;190:224-231.
76. Achenbach S, Daniel WG. Computed tomography of the heart. In: Libby P, Bonow RO, Mann DL, et al, eds. Braunwald's Heard Disease: A Textbook of Cardiovascular Medicine, 8th ed. Philadelphia, PA: Elsevier; 2008:415-435.
77. Greenland P, Bonow RO, Brundage BH, et al. ACCF/AHA 2007 Clinical Expert Consensus Document on Coronary Artery Calcium Scoring by Computed Tomography in Global Cardiovascular Risk Assessment and in Evaluation of Patients With Chest Pain. Circulation 2007;115: 402-426.
78. Hoffmann MHK, Shi H, Schmitz BL, et al. Noninvasive coronary angiography with multislice computed tomography. JAMA 2005;293:2471-2478.
79. Goldstein JA, Gallagher MJ, O'Neill WW, et al. A randomized controlled trial of multi-slice coronary computed tomography for evaluation of acute chest pain. J Am Coll Cardiol 2007;49:863-871.
80. Newman DH. Computed tomographic angiography for low risk chest pain: seeking passage. Ann Emerg Med 2009;53: 305-306.
81. Hollander JE, Chang AM, Shofer FS, et al. Coronary computed tomographic angiography for rapid discharge of low-risk patients with potential acute coronary syndromes. Ann Emerg Med 2009;53:295-304.
82. Takakuwa KM, Halpern EJ. Evaluation of a "triple rule-out" coronary CT angiography protocol: Use of 64-section CT in low-to-moderate risk emergency department patients suspected of having acute coronary syndrome. Radiology 2008;248:438-446.
83. Nair D, Carrigan TP, Curtin RJ, et al. Association of coronary atherosclerosis detected by multislice computed tomography and traditional risk-factor assessment. Am J Cardiol 2008;102:316-320.
84. Raff GL, Gallagher MJ, O'Neill WW, et al. Diagnostic accuracy of noninvasive coronary angiography using 64-slice spiral computed tomography. J Am Coll Cardiol 2005;46:552-557.
85. Harish A, Khatari P, Privadarshini H, et al. Accuracy of 64-slice coronary CT angiography in predicting percentage diameter stenosis. Indian Heart J 2008;60: 296-301.
86. Bamberg F, Schlett CL, Truong QA, et al. Presence and extent of coronary artery disease by computed tomography and risk for acute coronary syndrome in cocaine users among patients with chest pain. Am J Cardiol 2009;103:620-625.
87. Lee PW, Pellikka PA. Can computed tomography coronary angiography replace noninvasive functional stress testing? Am J Med 2008;121:358-359.
88. 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.
89. Hendel RC, Patel MR, Kramer CM, et al. ACCF/ACR/SCCT/SCMR/ASNC/NASCI/SCAI/SIR 2006 appropriateness criteria for cardiac computed tomography and cardiac magnetic resonance imaging: A report of the American College of Cardiology Foundation Quality Strategic Directions Committee Appropriateness Criteria Working Group, American College of Radiology, Society of Cardiovascular Computed Tomography, Society for Cardiovascular Magnetic Resonance, American Society of Nuclear Cardiology, North American Society for Cardiac Imaging, Society for Cardiovascular Angiography and Interventions, and Society of Interventional Radiology. J Am Coll Cardiol 2006;48:1475-1497.
90. Bastarrika G, Shyan Y, Rizsics B, et al. Coronary CT angiography: Applications. Radiol Clin North Am 2009;47:91-107.
I remember when the Goldman myocardial infarction (MI) algorithm came out, quickly followed by a seven-button hand-held "calculator." It promised to reduce all decision-making regarding ED chest pain patients to seven yes or no questions. But when you looked into the mathematics, if you answered no or negative to all of the questions, it indicated a 4% chance of acute cardiac ischemia. So, what would you do with this information? Could you tell the patient that there was only a 4% chance of a heart attack, so it was OK to go home?You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content