Expert Consensus Document on Cardiovascular Magnetic Resonance Imaging

The purpose of this expert consensus document is to inform practitioners and other interested parties of the opinion of experts concerning evolving areas of clinical practice and/or technologies that are widely available or new to the practice community. Topics are chosen for coverage because the evidence base or the experience with technology and/or clinical practice is not considered sufficiently well developed to be evaluated by the formal American College of Cardiology Foundation/American Heart Association (ACCF/AHA) practice guidelines process.

The goal of the expert consensus document should be regarded as the best attempt of the ACCF and document cosponsors to inform and guide clinical practice in areas where rigorous evidence may not be available or the evidence to date is not widely accepted. To that end, this document on cardiovascular magnetic resonance imaging (CMR) is timely and appropriate.

The CMR document is comprehensive and provides an enthusiastic recommendation for CMR and its multiple capabilities. The bibliography is comprehensive and the document provides a useful summary of the applications of magnetic resonance (MR) imaging in cardiovascular disease. A strength of the document is its comprehensive coverage. Several sections of the document are particularly well written and provide a useful guide to clinical practice.

In several clinical applications, the longstanding, widespread use of CMR makes expert consensus relatively straightforward. These include the use of CMR for the diagnosis of congenital heart disease, the evaluation of cardiac masses, pericardial disease, pulmonary venous anatomy, and congenital and acquired diseases of the thoracic aorta. Ideal for inherited or acquired diseases of the thoracic and abdominal aorta, both classic contrast-enhanced magnetic resonance angiography (MRA) or steady-state free precession (SSFP) (white-blood flow techniques), may be used. Velocity-sensitive algorithms may clearly identify the hemodynamic and perfusion-related peculiarities of acute and chronic aortic dissection. CMR techniques are particularly useful in surveillance of the natural course of aortic aneurysms and dissections and follow-up after surgical or endovascular repair. The widespread use of cine MR to assess wall motion and the presence of valvar regurgitation are again areas where CMR has been demonstrated to be a useful clinical tool providing pertinent information.

The consortium clearly provides the indication and appropriateness criteria for using CMR in ischemic heart disease, and clearly mentions the settings in which CMR is inappropriate to use. The document also provides helpful information in newer areas, such as the role of CMR in the evaluation of myocardial perfusion, the evaluation of heart failure and the use of CMR for the detection of the physiologic consequences of myocardial ischemia. The opinions regarding the role of CMR in the evaluation of myocardial scar should be of particular interest to clinicians. It’s value in the evaluation of nonischemic cardiomyopathy and myocarditis is well presented. The opinions regarding the application of CMR in the diagnosis of carotid artery disease, and peripheral vascular disease are supported by a reasonable body of evidence. The opinions regarding MR safety succinctly reflect and represent current opinion.

A criticism of the document is its tendency to present applications with less widespread use with the same level of enthusiasm as areas where CMR has widespread utility. These areas include the use of CMR for assessment of coronary atherosclerotic disease. Although CMR is widely used for determination of anomalous origin of the coronary arteries, there are relatively few centers where coronary CMR is performed, and considerable research needs to be done before coronary artery MRI is a reasonable tool for the complete evaluation of the coronary arteries and accurate assessment of coronary artery bypass graft status.

The section on phase-contrast blood flow measurement does not put into perspective that, although the technique is widely used, there is not a large body of literature addressing the frequent lack of correlation with echocardiographic and catheterization data. The use of phase contrast techniques to measure the hemodynamic significance of luminal compromise in the carotid and renal arteries requires further validation. Also requiring further validation is the use of CMR to detect increased myocardial water with T2-weighted imaging. Similarly, the consensus regarding use of CMR for measurement of left ventricular strain and myocardial tissue velocity measurements may lead some readers to assume that these MR measurements can be readily obtained and interpreted, which is not the case outside the research arena. The same is largely true for CMR plaque characterization, which remains an area of active research. An application not covered in the document is the use of CMR to assess central vein patency before line placement, and for the assessment of caval, portal, and peripheral vein patency in patients with suspected stenosis or thrombosis.

It should be acknowledged that in some vascular beds, noncontrast MRA techniques, which avoid exposure to gadolinium and its potentially nephrotoxic effects, may be useful. Time of flight and three dimensional (3D), noncontrast, electrocardiogram-gated and respiratory-navigated SSFP techniques may be employed. Both techniques allow 3D reconstructions.

The section on cardiac MR safety is critical and highly informative, comprehensively addressing the issue of CMR safety. Detailed information is provided on a wide variety of implanted devices with different ferromagnetic properties such as vascular stents, aortic stent graft from different manufacturers, intracardiac devices, coils leads, and aggregates. The information provided regarding coronary artery and peripheral vascular stents is well balanced. Similarly, the section on aortic stent grafts presents an evenhanded recommendation as does the section on intracardiac devices and inferior vena cava filters. Advantages such as high reproducibility and precision of cardiac functional parameters without ionizing radiation are important; however, considering that advanced heart failure often implies metallic device therapy including cardiac resynchronization therapy or implantable cardioverter defibrillator implants, surveillance and follow-up with CMR is inherently problematic. This is a limitation that needs to be addressed and activities to develop MR-compatible devices are not mentioned.[1]

Also, the application of 3 Tesla MR for the assessment of large vessel wall components and/or left atrium architecture is emerging and deserves addressing in the context of identifying patients with progressive atherosclerosis or with atrial fibrillation before attempted radiofrequency or cryoablation or pulmonary vein isolations.

The aspects of toxicity induced by gadolinium, such as nephrogenic systemic fibrosis (NSF), have been addressed and cautioned. Considering that the risk of NSF may climb up to 5% in end-stage renal disease, the use of contrast-enhanced MR should be clearly discouraged in this population rather than weighted against alternative use of possibly safer macrocyclic chelates or just avoiding a second exposure. Finally, the problem of teratogenicity and unknown risk of MRI during pregnancy is not discussed. Considering the importance of this aspect, though in the light of only scarce data, a word of caution is still warranted with a recommendation that careful assessment of risks versus benefits of MRI in the pregnant patient and fetus be done with preferential utilization of other imaging techniques, such as ultrasound, when possible. Because the effects of gadolinium containing contrast agents on the developing fetus are largely unknown, current consensus is avoidance of the administration of gadolinium-containing contrast agents in pregnant patients except in exceptional circumstances.

Overall, this is a comprehensive document that covers the widespread applications of CMR, showing CMR to be an extremely useful imaging tool for the assessment of cardiovascular disease. A leading manufacturer of MRI equipment has estimated that more than 60 million diagnostic MRI procedures are performed worldwide each year. In 2007, 27.5 million MRI procedures were performed the United States alone. CMRI studies constitute a sizeable portion of these imaging studies. With the large number of procedures performed, it is surprising that there is a relative lack of randomized controlled trials or large trials comparing the efficacy of CMR to other imaging techniques and validating hemodynamic measurements obtained with CMR.

However, when viewed in the context of clinical imaging, this circumstance is quite typical of what occurs with a new imaging procedure. If the imaging procedure provides images of such clarity that there is little confusion as to what one is seeing, and the clinical question answered with little patient risk, the imaging test will be ordered regardless of whether results from randomized trials or large studies are available; and utilization will outstrip scientific verification. If one can see a tree clearly, one does not need a randomized trial to determine that the object is a tree. If however, one cannot see the tree clearly and is confronted with a blurred, indistinct object, then tests need to be performed to determine the probability that the blurred object is indeed a tree and how likely the object is to be a tree. Such may explain the situation with CMR where, for most imaging procedures, images of high clarity are obtained. Validation of CMR for quantitative assessment of cardiovascular disease in some areas has lagged, as has validation in applications requiring high temporal or spatial resolution. It is in these latter areas that further research and verification are needed.