Defining the Role of Cardiac CT

There has always been a close connection between technological progress in cardiac imaging and improved management of patients with acquired and congenital heart disease. However, in the face of increasing costs and questions of long-term outcome benefit, we, as a society, have begun to question how we use new imaging technologies, seeking methodologies for decreasing health care costs, while maintaining high quality clinical services. To this end, the American College of Cardiology Foundation (ACCF), along with other specialty and subspecialty societies, conducted an appropriate use review of common scenarios where cardiac computed tomography (CT) is frequently considered. The 2010 Appropriate Use Criteria for Cardiac Computed Tomography [1] is an update of the report published in 2006.[2] Continued improvements in the technology of cardiac CT scanning (namely the integration of 64-detector scanners and dose reduction methods into clinical practice) and increased therapeutic options for patients with cardiovascular disease continue to contribute to the increasing utilization of noninvasive cardiac imaging. Clinicians ordering and specialists performing cardiac CT, as well as payers and patients, are interested in the appropriate use of this new technology. Inappropriate use may be potentially harmful to patients and costly; appropriate use should improve a clinical outcome. The use of appropriateness criteria represents a shift from addressing the question of potential risks and benefits of a particular treatment, to questioning the potential usefulness of a diagnostic test (i.e., cardiac CT) as a prelude to further treatment. The recommendations made in this report are based upon results of a modified Delphi exercise,[3] wherein in both evidence-based medicine (the literature) and experience-based judgment (embodied in a panel of radiologist and cardiologist cardiac CT specialists and non-CT specialists) are combined to reach a predefined level of consensus on the appropriateness (i.e., the expected incremental information, when combined with clinical judgment, exceeds expected negative consequences) of using cardiac CT in each of 93 clinical scenarios that reflect contemporary medical practice.

The scenarios for evaluation of the epicardial coronary arteries deemed appropriate for cardiac CT utilization include (1) in symptomatic patients with low and intermediate risk for coronary artery disease (CAD) with either an uninterpretable electrocardiogram (ECG) or the inability to exercise, (2) symptomatic individuals without known coronary heart disease and an uninterpretable ECG or the inability to exercise, normal ECG and cardiac biomarkers, uninterpretable ECG, or nondiagnostic ECG or equivocal cardiac biomarkers, (3) individuals with low or intermediate risk of CAD and new onset or newly diagnosed clinical heart failure with reduced left ventricular function, and (4) in patients with continued chest pain symptoms and a prior normal exercise test, or intermediate risk findings after exercise testing and a Duke Treadmill Score, or equivocal or normal stress imaging results, or discordant ECG exercise and imaging results. The report also describes as appropriate, the use of cardiac CT to evaluate graft patency after CABG (CABG) surgery, and for evaluation of (>3 mm) left main coronary artery stents. Appropriate use of cardiac CT includes evaluation of cardiac structure and function in adult patients with congenital heart disease, left ventricular function in patients following acute myocardial infarction or heart failure, with inadequate images from other noninvasive methods, quantitative evaluation of right ventricular function, and evaluation of right ventricular structure in suspected arrhythmogenic right ventricular dysplasia. Evaluation of native and prosthetic cardiac valves, or cardiac tumors is appropriate only if images from other noninvasive methods are inadequate. Cardiac CT is an appropriate means of evaluation of the pericardium, pulmonary vein anatomy before radiofrequency ablation for atrial fibrillation, the coronary veins before placement of a biventricular pacemaker, and the localization of CABGs and other retrosternal anatomy before reoperative chest or cardiac surgery

Use of noncontrast enhanced cardiac CT for evaluation of coronary calcium score was appropriate for the detection of CAD, and risk assessment in individuals with low risk but a family history of premature coronary heart disease, and with intermediate risk of CAD without known family history. If coronary calcium score is used as the criteria to then administer intravenous contrast material to perform a CT coronary arteriogram in symptomatic patients, then for scores up to 400 Agatston units, it is appropriate to perform cardiac CT.

Cardiac CT for the detection of CAD in symptomatic patients who have suffered a definite myocardial infarction, or symptomatic patients with a high pretest probability of CAD and an interpretable ECG and the ability to exercise was judged inappropriate. The use of cardiac CT for the detection of CAD in patients with a high risk of CAD before noncoronary cardiac surgery was also judged inappropriate. Furthermore, use of cardiac CT for risk assessment in the preoperative evaluation of patients without active cardiac disease for either low risk surgery irrespective of functional capacity, or for intermediate risk surgery or vascular surgery with functional capacity >4 METS, in the absence of clinical risk predictors, or if asymptomatic <1 year after a normal coronary arteriogram, stress test, or coronary revascularization, is inappropriate. The use of cardiac CT for the detection of CAD in patients with low- or high-risk findings on prior exercise testing and Duke Treadmill Score, or evidence of moderate or severe ischemia on prior stress imaging study was judged inappropriate. The use of cardiac CT for risk assessment in symptomatic patients who have received a coronary stent <3 mm in diameter, or in asymptomatic patients <5 years after CABG surgery, in whom a <3-mm diameter stent was placed, or in whom a >3-mm stent was placed >2 years ago, are all deemed inappropriate. Finally, the use of cardiac CT for the initial evaluation of left ventricular function after acute myocardial infarction, or in heart failure patients, or as the initial means of evaluating a patient suspected of an intracardiac mass or thrombus, is inappropriate.

Scenarios in which the report finds the use of cardiac CT to be uncertain include the detection of CAD (1) in symptomatic patients without known history of heart disease if they possess a high pretest probability of CAD, or low pretest probability but have an interpretable ECG and are able to exercise, (2) if there is persistent ECG ST-segment elevation following exclusion of acute myocardial infarction, or (3) in acute chest pain of uncertain cause (the "triple rule out"). In asymptomatic individuals with a high risk of CAD, obtaining a coronary calcium score or performing a coronary CT angiography is of uncertain appropriateness. In addition, obtaining a repeat coronary calcium score in asymptomatic individuals >5 years after a zero score or the use of cardiac CT to evaluate the coronary arteries after cardiac transplantation remain of uncertain appropriateness.

In patients with new onset or newly diagnosed heart failure, the use of cardiac CT for the detection of CAD in patients with high risk for CAD and reduced left ventricular function, or in patients with any level of pretest risk of CAD and normal left ventricular function is of uncertain appropriateness. The appropriateness of cardiac CT for the evaluation of CAD in individuals with low risk of CAD before non-coronary artery surgery is uncertain. Cardiac CT for the detection of CAD in patients with nonsustained ventricular tachycardia and syncope remain uncertain. The appropriateness of cardiac CT for the evaluation of elevated troponin of uncertain clinical significance in asymptomatic patients or without evidence of acute coronary syndrome is uncertain. The use of cardiac CT for the detection of CAD after evidence of mild ischemia on stress imaging, or in individuals with new or worsening symptoms with a previous abnormal stress imaging study are of uncertain appropriateness. Appropriateness of the use of cardiac CT for preoperative risk assessment before performing intermediate risk noncardiac surgery or vascular surgery in individuals with functional capacity <4 METS in uncertain. Risk assessment by cardiac CT in symptomatic individuals after revascularization (with a stent >3 mm or CABG), or in asymptomatic individuals >5 years after CABG, or >2 years after >3-mm stent placement remain uncertain. If a coronary calcium score is used as the criteria to administer intravenous contrast material to perform a CT coronary arteriogram in asymptomatic patients, the appropriateness of examination in patients with calcium scores >400 Agatston units is uncertain.

This report provides guidance for improving utilization of cardiac CT. The expanded number of clinical scenarios evaluated in the report provides support for rational utilization (patient selection) of cardiac CT in adult patients with acquired and congenital heart disease. The best use of cardiac CT is generally in symptomatic individuals with low or intermediate pretest risk of CAD, and in whom prior exercise or imaging was suboptimal. The same imprimatur applies to cardiac CT for the evaluation of adult patients with congenital heart disease or cardiac masses; if a conventional means of examination is suboptimal or unobtainable, then cardiac CT is appropriate. Thus, cardiac CT is often complementary to conventional noninvasive imaging and fills many unique clinical niches. This is similar to the role of cardiac magnetic resonance imaging: unique strengths and specialized niche utility.

The scenarios reported to be of uncertain appropriateness are worth commenting upon as well. Many of these scenarios involve symptomatic or asymptomatic individuals with high pretest CAD risk or positive prior cardiac test. As the authors point out, it is not unreasonable to use cardiac CT in these circumstances. However, there is simply limited data to encourage or advocate cardiac CT in these scenarios. Thus, these scenarios virtually define areas for future clinical investigation and subsequent peer review.

Cardiac CT has evolved over the past 3 decades to assume an important place in clinical practice. Increasing clinical experience, the reporting of multicenter clinical trials, and, of course, the lowering of patient radiation exposure continue to support efforts to optimize clinical utilization. I look forward to the next report from the committee, and to continued integration of this exciting imaging modality.