| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
EDUCATION EXHIBIT |
Professor Emeritus of Radiology, Stanford University School of Medicine, Stanford, California
The possibility of reliably performing noninvasive high-resolution imaging of the coronary arteries is at hand with the advent of 16–detector row CT (1,2). Because this technology is rapidly spreading and is now widely available, it is likely that requests for coronary CT arteriography will become increasingly frequent. Furthermore, most manufacturers have, or will soon have, CT scanners on the market that incorporate 32 or 64 rows of detectors, which will further improve the reliability and resolution of coronary CT arteriography. Therefore, radiologists should be prepared to interpret these studies, be aware of potential drawbacks, and know how to evaluate and eliminate confusing artifacts.
The three major stumbling blocks to noninvasive imaging of the coronary arteries are their small size (<1–3 mm), the fact that they are almost constantly in motion during the cardiac cycle, and their undulating distribution in three dimensions over a large volume. Ideally, image acquisition would include the entire volume of the heart and would occur virtually instantaneously—that is, in less than 50 msec, and preferably in less than 10 msec. This goal will not be achieved with present technology but may occur in the future with the advent of flat-panel devices with hundreds or thousands of detectors. In the meanwhile, postprocessing of axial images to create three-dimensional MIP, MPR, or shaded-surface-display images requires compiling images obtained from different cardiac cycles or different portions of the same cycle. However, doing so may lead to misregistration of images, creating step artifacts or pseudostenoses. Furthermore, the presence of calcification in the wall of the artery, implanted metallic stents, or adjacent metallic surgical clips can lead to blooming artifacts that impair visualization of the arterial wall.
In the preceding article, Nakanishi et al (3) systematically describe the artifacts commonly seen at coronary 16–detector row CT, identify their causes, and propose ways to minimize or mitigate them. These artifacts result primarily from cardiac pulsation, respiratory motion, and arrhythmias but can often be reduced by selecting a different phase of the cardiac cycle for imaging, slowing the heart rate with a ß-blocker, and carefully instructing the patient on how to suspend respiration for the duration of imaging. These techniques, if applied assiduously, will alleviate many or most postprocessing artifacts.
Partial volume averaging effect and streak artifacts due to high-attenuation entities are more difficult to eliminate. With each succeeding generation of scanners, however, resolution has improved significantly, and it is likely that 64–detector row CT will soon provide diagnostic images of vessels containing intraluminal stents or mural calcification (4). I remain skeptical that an accurate measurement of the lumen of a vessel 1 mm (1,000 µ) in diameter will be achieved unless vessels on the order of 100 µ are visible. Current technology does not provide this level of resolution, but gross findings of stenosis that appear similar to findings at cineangiography are routinely observed.
More disturbing is the potential for misinterpretation by a radiologist who lacks sufficient knowledge of the normal variations in coronary artery anatomy to properly distinguish a variant from a diseased vessel. Because viable myocardium must receive an adequate blood supply to function normally, it is incumbent on the interpreter to define the path of blood flow to each region of the left ventricular myocardium. In the absence of adequate blood flow, corresponding changes in the myocardium, such as thinning or calcification of infarcted tissue, should be apparent. In some cases of disease, collateral channels are the source of blood flow, a variant that must be noted and reported. In other cases, the blood flow is via a vessel with an anomalous origin from the aorta, or possibly from a pulmonary artery. Identifying these anomalies is one of the tasks of the radiologist, and such findings need to be communicated to the cardiologist or surgeon, especially if an intervention is contemplated. The interventional cardiologist may find it difficult to catheterize a coronary artery with an anomalous origin unless informed as to its location, and the surgeon may need to take added precautions in instrumentation in a patient with an anomalous coronary artery originating from the aorta.
The indications for noninvasive coronary angiography with multi–detector row CT have not been defined. It is likely that patients with known coronary artery disease who present with symptoms will not be candidates for noninvasive imaging because they will probably require catheterization and a percutaneous coronary intervention. However, patients who have previously undergone percutaneous coronary intervention or coronary artery bypass surgery and present with recurrent symptoms may be ideal candidates for noninvasive imaging. Coronary bypass graft patency is readily detected with multi–detector row CT (5), but stent patency is more difficult to determine because of beam-hardening artifacts (4). Patients who present to the emergency department with a chest pain syndrome but who do not have classic T-wave signs of an acute infarct or serum bio-markers for cardiac damage might benefit from noninvasive imaging, since as many as 50% of these patients will prove not to have coronary ischemia as the cause of chest pain (6). Multi–detector row CT is useful in this situation to rule out coronary artery disease as well as to explore other possible causes for chest pain, such as pulmonary embolus, airspace disease, or pleural disease. Patients with equivocal screening tests—particularly stress tests performed with either ECG alone, radioactive isotopes, or echocardiography—may also be candidates for noninvasive imaging because these tests frequently demonstrate false-positive findings.
The most controversial area for noninvasive coronary angiography is the screening of asymptomatic individuals. The majority of patients who present with a myocardial infarct were previously asymptomatic and had no prior warning signs; therefore, it seems prudent to try to identify asymptomatic patients who are at high risk for a coronary event. Some experts advocate screening of intermediate- and high-risk patients (defined according to the Framingham risk score) using coronary calcium determination to further stratify such cases and to determine a more optimal treatment plan to lower the risk of a future coronary event (7,8). At the present time, there are no data to suggest that adding noninvasive coronary angiography to coronary calcium determination enhances the ability to modify the treatment regimen (9). The temptation to perform a percutaneous coronary intervention or coronary artery bypass surgery in patients who are asymptomatic should be resisted, since there are no symptoms to relieve and the data suggest that life expectancy is not increased, except in patients with disease of the left main coronary artery or multivessel disease with compromised ventricular function (10).
I encourage radiologists to familiarize themselves with coronary multi–detector row CT angiography, coronary calcium determination, and other aspects of noninvasive cardiac imaging, such as myocardial perfusion, which is an ideal area for positron emission tomography or CT. Cardiac magnetic resonance imaging has certain advantages over multi–detector row CT in depicting and measuring myocardial blood flow and regional wall motion, but noninvasive coronary angiography is currently most reliably performed with multi–detector row CT. Future technical improvements in both CT and magnetic resonance imaging will help determine which modality is ultimately superior in this setting.
 |
References |
|---|
 Top References |
|---|
Related Article
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOGRAPHICS | RADIOLOGY | RSNA JOURNALS ONLINE |
