| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reviews |
1 From the Department of Diagnostic Radiology (C.S.W.) and Division of Emergency Medicine, Department of Surgery (D.K.), University of Maryland School of Medicine, 22 S Greene St, Baltimore, MD 21201. Received August 26, 2006; revision requested October 27; revision received December 15; accepted February 1, 2007; final version accepted March 16; final review and update by C.S.W. May 31. C.S.W. receives research support from Philips Medical Systems. Address correspondence to C.S.W. (e-mail: cwhite{at}umm.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONINITIAL ASSESSMENTRADIONUCLIDE PERFUSION IMAGING...CT SCANNINGCHALLENGESFUTURE DIRECTIONSESSENTIALSReferences |
|---|
© RSNA, 2007
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONINITIAL ASSESSMENTRADIONUCLIDE PERFUSION IMAGING...CT SCANNINGCHALLENGESFUTURE DIRECTIONSESSENTIALSReferences |
|---|
The extent of the challenge is exemplified by the need to separate serious causes of chest pain, such as angina, pulmonary embolism, and aortic dissection, from less serious ones. Perhaps the most important diagnosis to establish is that of coronary artery disease, which typically manifests as stable angina and acute coronary syndrome (ACS). ACS comprises the most serious forms of symptomatic coronary artery disease, including acute myocardial infarction, nontransmural myocardial infarction, and unstable angina (2).
It is evident in the rate of incorrect triage in the ED that no current paradigm is successful in identifying all patients with serious chest pain. In a study (3) of more than 10 000 ED patients with chest pain and related symptoms, 19 (2.1%) of 889 patients who were ultimately proved to have acute myocardial infarction were discharged inappropriately. Moreover, 2.3% of patients with unstable angina were mistakenly discharged. Other studies have indicated higher rates of incorrect diagnosis, and this challenge is not limited to the United States (4–6). Moreover, many patients who are proved not to have serious causes of chest pain are admitted unnecessarily for observation and further investigation, causing at least $8 billion in excess costs to be incurred in the United States each year (7).
|
INITIAL ASSESSMENT |
|---|
|
TOPABSTRACTINTRODUCTIONINITIAL ASSESSMENTRADIONUCLIDE PERFUSION IMAGING...CT SCANNINGCHALLENGESFUTURE DIRECTIONSESSENTIALSReferences |
|---|
Notwithstanding its value, the conventional triad has multiple limitations. As many as 20% of patients with ACS have chest pain that is atypical, and some patients who are ultimately proved to have myocardial infarction have no chest pain at all. In up to 10% of patients with myocardial infarction, the ECG result may be normal or show only nonspecific findings. Cardiac troponins have high specificity for myocardial infarction but low sensitivity in the initial few hours after onset of chest pain (8–11).
Patients with chest pain can be grouped at a first approximation into one of three primary groups. The first group has clear evidence of ACS as determined by means of clinical examination, ECG, or cardiac biomarkers. In this setting, patients are typically admitted expeditiously for coronary angiography and intervention, if necessary, in accordance with guidelines of the American College of Cardiology and the American Heart Association (12). The second group consists of patients with symptoms that are clearly less serious. An example would be a 25-year-old individual with chest pain that is reproducible with palpation.
The third group provides the greatest challenge. This group comprises patients who often are in their 4th through 6th decades of life, with chest pain that is deemed indeterminate at initial work-up. Patients may have few if any risk factors and a clinical history that is atypical, with chest pain that is of a stabbing or sharp variety rather than the chest pressure or tightness that is typically associated with coronary artery disease. Likewise, the ECG results may show only nonspecific T wave changes. The biomarkers in these patients are normal, at least initially. It is this group that constitutes a large percentage of the patients, 28% in one study, who present to the ED and who require further diagnostic evaluation (13). Imaging has a large potential to affect the triage decisions made in this indeterminate group.
A variety of imaging techniques are available to assess patients in the ED with indeterminate chest pain. Routine chest radiographs may show causes of chest pain, such as pneumonia, pneumothorax, or rib fracture and are typically included in the evaluation of most patients presenting to the ED with chest pain. In rare situations, ACS may be suggested by the presence of coronary arterial calcification or evidence of congestive heart failure (14,15), but typically the chest radiograph is normal. Radionuclide perfusion imaging and echocardiography are also used frequently in this population and may provide valuable information. Magnetic resonance (MR) imaging has also been suggested as an option (16). Recently, CT scanning has been proposed as a method to help triage patients with indeterminate chest pain.
|
RADIONUCLIDE PERFUSION IMAGING AND ECHOCARDIOGRAPHY |
|---|
|
TOPABSTRACTINTRODUCTIONINITIAL ASSESSMENTRADIONUCLIDE PERFUSION IMAGING...CT SCANNINGCHALLENGESFUTURE DIRECTIONSESSENTIALSReferences |
|---|
Findings of multiple studies have shown sensitivity ranging from 90% to 100%, specificity from 60% to 78%, and negative predictive value from 97% to 100% at radionuclide perfusion imaging for ACS if single photon emission computed tomography (SPECT) imaging is used (18–24). The addition of stress imaging in stable patients may improve diagnostic accuracy (25). In addition, radionuclide perfusion imaging has been shown to have prognostic value and permits risk stratification of patients for future cardiac events (1).
Notwithstanding these benefits, it is clear that radionuclide perfusion imaging has limitations. The nuclear medicine department is usually not located in or near the ED, which requires the patient to be moved. During off hours, the study may not be available in a timely fashion because it requires that a nuclear medicine technologist be brought in on-call. Finally, the focus of the study is the assessment of coronary artery disease, and other causes of chest pain cannot be evaluated.
Echocardiography also may be used to evaluate the ED patient at low to moderate risk of ACS. Findings indicative of myocardial ischemia include segmental wall motion abnormalities and reduced ejection fraction. In a study comparing echocardiography and radionuclide perfusion imaging, Kontos et al (26) found similar sensitivity and specificity. An additional advantage of echocardiography is portability (26). Drawbacks include limited availability during off hours and false-negative results in patients with small myocardial infarctions or unstable angina. Early experience suggests that myocardial contrast material–enhanced echocardiography may improve the diagnostic accuracy of this technique (27,28).
|
CT SCANNING |
|---|
|
TOPABSTRACTINTRODUCTIONINITIAL ASSESSMENTRADIONUCLIDE PERFUSION IMAGING...CT SCANNINGCHALLENGESFUTURE DIRECTIONSESSENTIALSReferences |
|---|
The major cardiac application for which electron-beam CT has been used is coronary calcium scoring. Coronary arterial calcification occurs as part of the spectrum of coronary atherosclerosis and is considered highly specific for its presence (30). A relationship exists between the presence and amount of coronary artery calcification and coronary artery stenosis, but this relationship is not linear (31). Likewise, there is a correlation between the presence and amount of coronary artery calcification and the likelihood of a substantial coronary event, often defined as the development of ACS, sudden death, or the requirement for coronary revascularization by using percutaneous intervention or coronary artery bypass graft placement (32). Notwithstanding these relationships, it is clear that the absence of coronary calcium does not exclude the presence of coronary atherosclerosis (33,34).
The electron-beam CT calcium scoring protocol typically consists of 30–40 3-mm-thick sections obtained through the heart during one or two breath holds. ECG triggering is typically employed in a prospective manner at 80% of the R-R interval. Software that highlights pixels with an attenuation of greater than 130 HU, the conventional threshold for CT-depicted calcium, is applied to all images. Regions of interest are placed around highlighted areas that correspond to the coronary arteries. The calcium score is a measure of the number and density of the highlighted pixels and can be expressed as an Agatston, volume, or mass score (35). Whereas the largest amount of experience exists with the Agatston score, the last two methods are considered most appropriate for volumetric imaging (36).
Electron-beam CT has been used primarily in asymptomatic patients to further define coronary risk. However, authors of three studies have investigated the value of calcium scoring at electron-beam CT to triage patients who present to the ED with chest pain (37–39). Laudon et al (37) assessed 105 patients with possible ACS. Inclusion criteria were normal cardiac enzymes and a nondiagnostic ECG result; patients with known coronary artery disease were excluded. All qualifying patients underwent electron-beam CT, and a nonzero calcium score was considered positive for ACS. The ED physician caring for the patient was blinded to the electron-beam CT results. A final determination was made on the basis of clinical impression and conventional testing. The overall sensitivity, specificity, and negative predictive value for a positive finding at electron-beam CT were 100%, 63%, and 100%, respectively. No patient with a negative result at electron-beam CT had a cardiac event at 4-month follow-up. The authors concluded that patients with a negative result at electron-beam CT may be safely discharged. In an investigation with similar design, McLaughlin et al (38) studied 134 patients with indeterminate ECG results and normal cardiac enzyme findings by using calcium scoring. Among 86 patients with a positive finding at electron-beam CT (defined as Agatston score of >1), seven (8%) had cardiac events. Only one (2%) of 48 individuals with a negative CT finding had an event. Like Laudon et al, these authors concluded that electron-beam CT may be valuable in triage for chest pain. Similar recommendations were made in a prospective observational study by Georgiou et al (39), in which 192 patients admitted to the ED who underwent electron-beam CT were followed up for an average of 50 months.
Notwithstanding these promising results, electron-beam CT images are noisy, and the test has limited applicability to other parts of the body. In addition to the cost of the scanner, further technology development has not occurred. In recent years, electron-beam CT has largely been supplanted by multidetector CT.
Multidetector CT
The commercial availability of ECG-gated four–detector row CT beginning in the late 1990s ushered in the current era of cardiac CT. This development raised the prospect that coronary CT angiography might provide a noninvasive alternative to coronary angiography, which for decades has been considered the standard of reference to evaluate coronary disease. Multiple investigations with this earlier generation of multidetector scanners demonstrated the potential of multidetector CT in comparison with coronary angiography, although the results were limited by the large number of nonevaluable segments at coronary CT angiography (40–43). In recent years, there has been further substantial advance in the capability of multidetector CT.
The current generation of scanners consists of 64 detector rows with an isotropic spatial resolution on the order of 0.5 mm and a temporal resolution of 200 msec or less. This spatial resolution is about one-third of that achievable by means of coronary angiography and typically permits visualization of vessels larger than 2 mm. However, current temporal resolution at coronary CT angiography is substantially inferior to that at coronary angiography. To decrease motion artifact, a heart rate of less than 65 beats per minute is desirable. In patients with higher heart rates, β-blockers are often administered to reduce the rate. Nitroglycerin is often used to vasodilate the coronary arteries, provided that the patient has no contraindications, such as hypotension.
A rapid infusion of intravenous contrast material, 5–6 mL/sec, is used with a bolus triggering technique or test injection. Although 10 equally spaced images (phases) throughout the cardiac cycle are often reconstructed, images obtained during diastole typically provide the best image quality of the coronary arteries and some selectivity can be employed as to which images to reconstruct. With 64–detector row CT, sensitivity and specificity rates of greater than 80% have been reported for coronary CT angiography versus invasive angiography, with fewer nonevaluable segments than with earlier-generation multidetector CT scanners (44–46).
Apart from assessment of the coronary arteries, cardiac CT angiography can provide important information about myocardial function. Left ventricular wall motion is readily evaluated if a series of images (phases) is obtained throughout the cardiac cycle. End-diastolic and end-systolic images enable calculation of a left ventricular ejection fraction, with good correlation (r = 0.89–0.97) compared with other techniques, particularly MR imaging (47–49). An abnormality of myocardial perfusion may be visualized as an area of low attenuation in the left ventricular myocardium after contrast material enhancement (50,51).
In addition to coronary CT angiography, multidetector CT is capable of being used to measure coronary artery calcium with a protocol analogous, but not identical, to electron-beam CT (52). Although calcium scoring is usually performed with the Agatston method, as noted previously, the best correlation of calcium scores between multidetector CT and electron-beam CT is achieved by using volumetric or mass scoring methods, which account for the volumetric nature of current CT data (36).
Multidetector CT in the ED
In addition to improved technology, the increasing placement of CT scanners in or near the ED suite is another factor that has heightened the attractiveness of using multidetector CT to evaluate acute chest pain. Multidetector CT currently is used routinely to assess many nonthoracic emergency indications, such as abdominal pain, headache, and trauma, as well as critical extracardiac thoracic disease, including pulmonary embolism and aortic dissection. Thus, ED physicians have become increasingly familiar with and have developed expectations regarding the capabilities of multidetector CT.
Multidetector CT has the potential of playing either of two roles in the triage of ED patients with chest pain of possible cardiac origin. In the first scenario, multidetector CT images are obtained as soon as possible after initial assessment, generally after patient history, physical examination, and ECG are completed and the first set of cardiac enzyme and serum creatinine levels are found to be normal. In this context, the results become a component of the decision to discharge the patient or admit the patient to the hospital. In an alternative scenario, the multidetector CT scan is acquired after a decision is made to admit the patient, and the findings assist in determining the type of monitoring and treatment that are indicated, such as telemetry, percutaneous intervention, or surgical revascularization.
Another unresolved issue in patients with chest pain in the ED is whether to perform dedicated coronary CT angiography or to scan the entire thorax. A dedicated coronary CT angiography examination performed by using the latest generation of multidetector CT scanner provides the best possible images of the coronary arteries and allows limited visualization of other structures that may be responsible for chest pain, such as the central pulmonary arteries (53). This protocol may prove valuable in patients who are judged to have non–life threatening causes of cardiac disease (eg, stable angina) after initial clinical evaluation and in whom other serious causes of chest pain have been excluded.
The alternative method of scanning has been termed the "triple rule-out" approach to designate the triad of coronary artery disease, pulmonary embolism, and aortic dissection (Figs 1 and 2). It has also been labeled the comprehensive or global assessment (54,55). Those investigators who favor a comprehensive approach invoke the well-documented ability of multidetector CT to provide a simultaneous assessment of multiple causes of chest pain (Fig 3). Whereas some patients present with chest pain symptoms characteristic of a particular entity, these investigators note that many other patients have clinical indicators that are nonspecific.
|
|
|
|
|
|
|
The role of coronary calcium assessment in the ED CT protocol is uncertain. If calcium is absent, one might suggest that further evaluation with CT angiography is unnecessary. However, the absence of calcium decreases the likelihood of significant coronary stenosis (>50% decrease in luminal diameter) but does not exclude it. Moreover, other causes of chest pain (such as pulmonary embolism) that require intravenous contrast material are not assessed if only calcium scoring is performed. Conversely, in patients with very high amounts of coronary calcification, coronary CT angiography is likely to be suboptimal due to artifacts from calcium but not necessarily nondiagnostic. The calcium score also may prove valuable to the ED physician in triaging patients with coronary disease (Table 2).
|
By using a triple rule-out (global assessment) protocol and 16–detector row CT, our group studied 69 patients who presented to the ED with indeterminate chest pain defined as low to intermediate risk of ACS with a normal or nonspecific ECG findings and normal initial cardiac enzyme results (55). Patients were imaged immediately after initial clinical evaluation was performed and a serum creatinine level was obtained. Because of lengthy postprocessing requirements at the time of the study, clinicians were not informed of the results of the coronary evaluation in real time. Prior to scanning, the ED physician was asked whether the CT scan would have been obtained for conventional clinical indications. The standard of reference for the cause of chest pain was determined in consensus by a cardiologist, an emergency physician, and a radiologist who reviewed the medical records for coronary events within 1 month of ED discharge, as well as for any correlative imaging studies such as radionuclide perfusion imaging, echocardiography, or coronary angiography. On the basis of this consensus, overall sensitivity, specificity, positive predictive value, and negative predictive value were 87%, 96%, 87%, and 96%, respectively, for multidetector CT in the evaluation of ED chest pain. In addition to 10 patients with significant coronary disease (>50% luminal stenosis), three noncoronary diagnoses (pulmonary embolism, pneumonia, pericarditis) were established by using the larger field-of-view triple rule-out protocol. Ultimately, acute myocardial infarction was diagnosed in only one patient, yet 29 (42%) of 69 patients required admission, nearly all for evaluation of unstable angina. Many of these patients were proved not to have coronary disease at CT angiography. On the basis of our survey of ED physicians prior to CT, 65% of the study patients would not have undergone CT scanning for conventional indications. Savino et al (54) recently reported promising preliminary results by using a global assessment protocol and 64-section CT in 23 patients who presented to the ED with chest pain. In that study, the finding of significant coronary stenosis at multidetector CT in each of eight patients with this finding was confirmed at invasive angiography.
More recently, Goldstein et al (62) performed a randomized trial in 197 patients with acute chest pain at low risk for ACS presenting to the ED and compared the safety, diagnostic efficiency, and cost of 64–detector row CT with the standard of care diagnostic evaluation. Physicians using multidetector CT were able to immediately exclude or identify coronary disease as the source of chest pain in 75% of patients. The remaining 25% of patients required stress testing because of intermediate severity lesions or nondiagnostic scans. The patients who underwent multidetector CT had a safety profile (no adverse coronary events for 6 months) equal to that of patients treated with the standard of care approach. Initial evaluation with multidetector CT reduced diagnostic time to an average of 3.4 hours compared with an average of 15 hours with a standard of care evaluation. Moreover, average cost per patient was lowered from $1872 to $1586 when the multidetector CT algorithm was used. Patients in the multidetector CT arm of the study also required fewer subsequent evaluations for recurrent chest pain.
The discussed studies indicate the feasibility of using CT angiography to assess chest pain in the ED and potentially lead to earlier triage. The most appropriate group for this application appears to be patients at low to intermediate risk for ACS. It appears likely that admissions for chest pain can be reduced if CT is incorporated into the protocol, owing to its high negative predictive value. However, it is important to recognize that the study results published to date consist of small numbers and are limited in scope because each represents the experience of a single center. There is no consensus as to the role of multidetector CT in the ED evaluation for chest pain relative to other imaging techniques, such as radionuclide perfusion imaging. A more definitive set of indications for the use of multidetector CT is necessary to minimize overutilization. It is also clear that use of multidetector CT will identify a group of patients with subclinical coronary disease (ie, disease that is not the cause of the chest pain). The appropriate follow-up of such patients with nonsignificant coronary atherosclerotic plaque remains a matter of debate. A separate area of uncertainty applies to patients with a positive CT angiogram (coronary artery stenosis > 50%). These patients typically undergo further noninvasive or invasive testing, but the precise course of further evaluation is unresolved. Finally, some serious causes of chest pain may not be amenable to diagnosis by means of multidetector CT (eg, electrolyte imbalance). A sample triage protocol based on a clinical consensus of the ED physicians and cardiologists at our institution is listed in Table 2.
|
CHALLENGES |
|---|
|
TOPABSTRACTINTRODUCTIONINITIAL ASSESSMENTRADIONUCLIDE PERFUSION IMAGING...CT SCANNINGCHALLENGESFUTURE DIRECTIONSESSENTIALSReferences |
|---|
A second important issue related to the generation of increased CT volume is the size of radiation dose to which patients are exposed in the course of a CT angiography examination. It is estimated that exposure during dedicated cardiac CT angiography exceeds 10 mSv for a single examination or at least triple the amount occurring from natural exposure on an annual basis (64). The use of a global protocol is associated with an even higher dose because of longer scanning length. Dose modulation or ECG pulsing, in which tube current is markedly reduced during less critical parts of the cardiac cycle, is one method that can decrease radiation exposure by as much as 40% (65). An additional consideration is that several of the current imaging studies used to evaluate chest pain in the ED, such as radionuclide perfusion imaging and coronary angiography, are also associated with substantial radiation doses. Some of the radiation burden might be mitigated if CT angiography proves to be an acceptable substitute for these techniques.
A third consideration is the economic impact of introducing CT angiography into the triage algorithm of ED chest pain. As suggested in our pilot study, the introduction of CT angiography to evaluate coronary chest pain will likely lead to its more frequent use. A substantial increase in volume will necessarily be accompanied by higher costs for CT. However, if other parts of the work-up can be obviated, and in particular, if admissions are decreased by the use of CT angiography, the net cost to the patient and third-party payer may remain stable or decrease.
We believe a rigorous clinical trial is necessary to determine the proper role of CT angiography in the evaluation of chest pain in the ED. Such a trial should compare the use of CT angiography with current triage methods, with a primary endpoint of efficacy, as measured by the prevalence of patient presentation with a serious cause of chest pain following discharge from the ED. For CT angiography to be effective, the rate of inappropriate discharge cannot be higher than that with conventional treatment. Other important endpoints are length of stay in the ED, rate of admission, and comparative cost.
|
FUTURE DIRECTIONS |
|---|
|
TOPABSTRACTINTRODUCTIONINITIAL ASSESSMENTRADIONUCLIDE PERFUSION IMAGING...CT SCANNINGCHALLENGESFUTURE DIRECTIONSESSENTIALSReferences |
|---|
In summary, initial investigations suggest that CT angiography has considerable potential to streamline chest pain evaluation in the ED, but further investigation is imperative to establish its precise role.
|
ESSENTIALS |
|---|
|
TOPABSTRACTINTRODUCTIONINITIAL ASSESSMENTRADIONUCLIDE PERFUSION IMAGING...CT SCANNINGCHALLENGESFUTURE DIRECTIONSESSENTIALSReferences |
|---|
|
FOOTNOTES |
|---|
Abbreviations: ACS = acute coronary syndrome • ECG = electrocardiography • ED = emergency department
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONINITIAL ASSESSMENTRADIONUCLIDE PERFUSION IMAGING...CT SCANNINGCHALLENGESFUTURE DIRECTIONSESSENTIALSReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. C. Hendel Is Computed Tomography Coronary Angiography the Most Accurate and Effective Noninvasive Imaging Tool to Evaluate Patients With Acute Chest Pain in the Emergency Department?: CT Coronary Angiography Is the Most Accurate and Effective Noninvasive Imaging Tool for Evaluating Patients Presenting With Chest Pain to the Emergency Department: Antagonist Viewpoint Circ Cardiovasc Imaging, May 1, 2009; 2(3): 264 - 275. [Full Text] [PDF]
|
|
|
|
 |
|
 A. S. Jaffe Key Issues in the Developing Synergism between Cardiovascular Imaging and Biomarkers Clin. Chem., September 1, 2008; 54(9): 1432 - 1442. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. M. Takakuwa and E. J. Halpern 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, August 1, 2008; 248(2): 438 - 446. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
