Advanced Imaging Technology in the Detection of Coronary Artery Disease: One Step at a Time

In 1958, while performing an ascending aortogram, the tip of Mason Sones’ catheter inadvertently cannulated the right coronary artery and injected contrast material. The patient survived and thus began a new paradigm of investigation for the detection of coronary artery disease. Decades later, coronary angiography is now performed annually in over 2 million people in the United States alone and has earned the status of the gold standard for the diagnosis of coronary artery disease. This technique has been limited principally by its invasive nature. To overcome this limitation, noninvasive testing has been applied and investigated to enhance its ability to predict the risk for future events and its ability to detect coronary artery disease. The most commonly used noninvasive tests (nuclear perfusion imaging and echocardiographic imaging coupled with pharmacologic or physiologic stress testing) have been incorporated by the American College of Cardiology and American Heart Association in the guidelines for management of patients with symptoms suggestive of acute coronary syndromes and stable angina. The advantage of these non-invasive tests is the fact that a negative predictive value is coupled with a good prognosis with respect to the occurrence of future events. Advanced imaging techniques (magnetic resonance imaging, position emission, and CT angiography) are now gaining prominence as the next generation non-invasive tests for the detection of coronary disease. In this month’s cover article Advanced Imaging’s Impact On The Cardiac Catheterization Lab Will Depend Upon Philosophy, Weaver discusses a scenario in which advanced imaging would substitute for diagnostic cath and be used as a screening tool, focusing on the overall impact on the cath lab. Overall, the author indicates that cath lab patient volumes will drop, but a greater percentage of patients entering the cath lab will undergo percutaneous interventions. However, for advanced imaging to become a significant feature or even a standard in the work-up of angina, a major shift in clinical practice will be required. While magnetic resonance imaging of the coronaries has improved immensely, it is believed that this technology will be limited by the relative lack of scanners and timeliness required to obtain the necessary images for each patient. Likewise, positron emission tomography is not widely available, making these two technologies restricted to the larger centers across the country. Thus, the true new player is really just CT angiography. The emergence of CT angiography incorporating 16- or 64-slice imagers has the potential to become a revolutionary new tool in the detection of coronary disease. As this technology is new, the positive and negative predictive value and/or long-term outcome data for patients undergoing CT angiography remain under study. However, Hoffman et al¹ did investigate the use of multislice CT angiography in 103 consecutive patients (all having intermediate to high risk for coronary artery disease and also scheduled to undergo coronary angiography). Impressively, they found that the sensitivities and specificities were above 90% and 75%, respectively, for the detection of significant angiographic coronary disease (greater than 50% left main coronary disease or greater than 70% stenosis in a major epicardial vessel). There were, however, a small number of coronary imaging sequences (6.4%) which could not be interpreted. The authors suggested that these findings supported the complementary role of CT angiography and invasive coronary angiography. One must keep in mind several issues which must be considered with the incorporation of new technologies in patient care scenarios, such as CT angiography. The first issue is the two factors involved in obtaining an accurate sensitivity and specificity for testing, namely the patient population being tested and the standard against which the test is compared. Hoffman (above) focused on patients at intermediate to high risk for having coronary disease and symptoms compatible with coronary disease. The patients were all scheduled to undergo diagnostic coronary angiography. One cannot assume that sensitivity and specificity would remain at the same values for screening asymptomatic patients or for those patients at lower risk for having obstructive coronary disease. As we have seen in the evaluation of other non-invasive testing, such as V/Q scans for the detection of pulmonary embolus in the PIOPED study², the true accuracy of the test is largely dependent on the patient's pretest probability of having disease (e.g., a pulmonary embolism). The patients with a lower likelihood of embolism did not benefit from V/Q scanning, as its accuracy in these patient dropped significantly compared to patients with higher likelihood of having a PE. This relationship applies to our understanding of the role of CT angiography in present-day practice. Patients with a high likelihood of coronary disease presenting with unstable angina often are brought directly to the cath lab. It is unlikely that CT angiography becomes an intermediary step in this process, as the coronary anatomy needs to be defined in these patients, and it is the cath lab which offers the potential for revascularization (i.e., both diagnostic and therapeutic benefit). Thus, the true impact of an imaging modality will likely be in its ability to perform in patients with, at most, an intermediate probability of disease (i.e., can it reduce the number of normal caths?). Secondly, current noninvasive testing modalities perform at their optimum when anatomic and physiologic data are combined. This is seen in nuclear perfusion imaging when changes in myocardial perfusion become more predictive when accompanied by LV wall motion abnormalities. Coronary angiography in itself does not predict the ischemic potential of fixed obstructive lesions or the likelihood that modest narrowings do produce chest pain. This observation can be best understood when comparing invasive angiography and fractional flow reserve (FFR) in patients with suspected ischemia-producing lesions. In a study by Fischer et al³, experienced interventional cardiologists were able to predict ischemia-producing lesions only in about 50% of cases, resulting in an 80% sensitivity but only 47% specificity for the prediction of ischemia produced by an epicardial stenosis. These findings were not improved when quantitative coronary angiography was utilized to analyze the epicardial lesions. CT angiography, like conventional coronary angiography, only produces anatomic data. The question then becomes, at what degree of coronary narrowing by CT angiography should patients be referred to the catheterization lab for further evaluation and potential revascularization? In our current practice environment, this threshold will likely be quite low, and thereby the overall effect on the catheterization lab could be a large increase in patients with intermediate coronary disease. How then, should we differentiate treatment in patients with angiographic (invasive or by CT) coronary lesions that are 40 to 70%? If these patients are to undergo a catheterization, then fractional flow reserve assessment of these intermediate lesions remains a proven next step to determine which lesions are properly approached with stenting.⁴ Combined anatomic and physiologic data will remain the standard to which we should hold ourselves in managing our patients. The quest for improved imaging technologies and diagnostic capability to diagnose coronary artery disease remains quintessential to our current practice. However, all new testing modalities must undergo rigorous testing and scrutiny that previous modalities have undertaken. In cardiology practice, we enjoy the privilege of a wealth of evidence to support the many clinical decisions that we must render. We must not lose sight of this requirement in light of the upcoming new or fancier testing modalities. I urge invasive and interventional cardiologists to embrace learning about new testing modalities such as CT angiography, and require rigorous testing protocols to be undertaken and presented before appropriate implementation of these tests in clinical practice. With accurate data, I have no doubt that CT angiography will indeed become an important complement to invasive coronary angiography and our patients will be much better off as a result. Email: limmj@slu.edu

1. Hoffmann MHK, Shi H, Schmitz BL, et al. noninvasive coronary angiography with multislice computed tomography. <i>JAMA</i> 2005; 293:2471-2478.<p>2. The PIOPED investigators. Value of ventilartion perfusion scan in acute pulmonary embolism. Results of the prospective investigation of pulmonary embolism diagnosis (PIOPED). <i>JAMA</i> 1990;263:2753-2759. </p><p>3. Fischer JJ, Samady H, McPherson JA, et al. Comparison between visual assessment and quantitative angiography versus fractional flow reserve for native coronary narrowings of moderate severity. <i>Am J Cardiol</i> 2002;90:210-215. </p><p>4. Bech GJ, DeBruyne B, Bonnier HJPM, et al. Long-term follow-up after deferral of percutaneous transluminal coronary angioplasty of intermediate stenosis on the basis of coronary pressure measurement. <i>J Am Coll Cardiol</i> 1998;31:841-847.</p>