Coronary MR Angiography: Can It Compete With Multidetector CT?

Just five years ago, the thought of cardiac cross-sectional imaging turned immediately toward the potential of MRI for evaluation of the heart. Equipment manufacturers provided new MRI scanners with fast gradients for this purpose and developed new pulse sequences. Advanced 3D workstations provided new software for analysis of cardiac function, and stress tests started to be performed in the MRI scanner. Everyone understood the superiority of MRI, compared with echocardiography. The key word in the preceding description remains the potential of MRI. Today, when we think of cardiac imaging, our thoughts move immediately toward cardiac multidetector CT (MDCT). These machines can combine hundreds of rows of detectors for imaging the heart in a fraction of a second; 3D images display spinning hearts and cutaway views of the coronary arteries with astounding clarity. Turf wars that were a concern with cardiac MRI never manifested themselves, but turf wars may potentially rage again with cardiac CT. What went wrong with MRI? Instead of something going wrong with MRI, there were sudden, tremendous advances in a single job done well by CT: noninvasive imaging of the coronary arteries. Disease of the coronary arteries kills more than a million people in the United States per year through acute myocardial infarction. Thus, coronary-artery visualization by any imaging modality becomes central to patient management. Today, patients with suspected coronary artery disease still undergo invasive radiographic cardiac catheterization. The likelihood of finding coronary-artery disease in any particular patient is relatively high: our diets are often unhealthy, 60% of us are overweight or obese, and many of us have hypertension and/or diabetes and exercise very little. Despite having some coronary disease, however, only about one in five patients who undergoes radiographic invasive cardiac catheterization has coronary disease that is treatable. Cardiologists clearly understand the disadvantages of this shotgun approach to the use of heart catheterization; they need an alternative that is noninvasive, safe, and rapid. MDCT fits this prescription exactly. Paradigm Shift We are currently in the midst of a major paradigm shift for coronary disease. It is similar to that seen a few years ago for diagnostic radiographic peripheral catheter angiography. Currently, no patient will receive a peripheral radiographic angiogram without first having undergone a noninvasive CT or MRI angiography examination. As should be the case, noninvasive tests represent the gateway to determine who should receive the invasive test. The next imminent gateway is the use of MDCT for coronary-artery disease. The success of MDCT coronary angiography for CT scanners with at least 64 rows of detectors is excellent: multiple studies report sensitivity and specificity rates of 90% to 95% for significant coronary-artery narrowing. These are much higher than rates for methods such as perfusion stress imaging of the myocardium, attesting to the excellent anatomic depiction of MDCT, compared with catheter angiography. The negative predictive value of MDCT is the percentage of cases shown as being without significant disease, confirmed when the catheter angiogram indicates that no disease is present. For MDCT, the negative predictive value is consistently greater than 95% in multiple studies. This indicates that there is very little significant disease missed by MDCT. With all of this good news for MDCT of the coronary arteries, is there anything that can possibly halt the progression of this technology? For the most part, the test is relatively simple, by the standards of invasive testing. On most MDCT scanners, image quality is improved after heart-rate reduction. A beta blocker can be administered, usually by mouth, for heart-rate control. For most patients, this is a safe procedure, with heart-failure patients being one of the more common exceptions. Radiation dose is a new concern of many in the medical community and the public. The impact of medical radiation on the health of the US population is still poorly understood. The only factor that is entirely clear is that less radiation is better. The radiation dose from cardiac testing never seemed to be an important topic prior to the adoption of MDCT. Invasive angiography with radiography administers about 5 mSv (and the average background radiation in the United States is about 3.6 mSv). Nuclear medicine stress testing can involve much higher doses of 15 to 20 mSv. Older-technology MDCT also exposes the patient to 15 to 20 mSv, so why did radiation dose become such an issue? Clearly, everyone understands the inherent potential of CT; the growth in CT imaging is 10% to 15% per year. Coronary CT has the potential to contribute to the medical radiation dose substantially. Fortunately, CT-scanner manufacturers are rapidly responding to this potential radiation crisis. The manufacturers have understood that this potential concern is not only bad for patient care, but also bad for the corporate bottom line. With the latest generation of 128- to 320-detector CT scanners, radiation doses can be 5 mSv or less per CT examination. The acceptance of the medical community and public of these lower radiation doses remains to be seen. MRI: What Happened? What about MRI? Has it completely fallen by the wayside? Can we now determine, in retrospect, what went wrong? In some respects, nothing really is wrong with MRI. It does a superb job of imaging myocardial function. It has established itself as the gold standard for function and for detection of myocardial scarring. MRI accurately measures flow in the heart or vessels, and MRI stress testing competes on a par with echocardiography or nuclear medicine tests. What about the coronary arteries? Herein lies the problem. The achievement crucial to CT was reliable imaging of the coronary arteries; this was not so for MRI. At least 15% of patients who have coronary-artery imaging with MRI have studies that are not interpretable due to motion or other artifacts. The spatial resolution is worse than that of CT, and the examination takes at least 30 minutes (in an uncomfortable scanner). sensitivity in detecting coronary disease is lower than for CT, at about 80% to 85%. The interesting feature of MRI is that coronary-artery imaging could be nearly on par with that of CT (except for longer scan times). Large scientific meetings, such as those of the International Society for Magnetic Resonance in Medicine, routinely have academic, scientific presentations showing advances in MRI far beyond what scanner manufacturers currently provide to their customers. Slow adoption and lack of aggressive research-and-development programs are largely the reasons for lack of progress in coronary imaging with MRI. Besides, with CT now so overwhelmingly good, is there a need for MRI of the coronary arteries? The answer to this question remains yes. Consider angiography of the peripheral vessels: Is there a need for both CT and MRI? To answer this question, one simply needs to think about one’s own health care. Would you rather have a CT scan with radiation and iodine administration or an MRI scan without injection of contrast and no radiation dose? Many of us (if we are not claustrophobic) would readily request the MRI scan if it worked as well as CT. For people with claustrophobia, we need larger, more comfortable, and quieter MRI scanners. All this technology is potentially available for MRI, but is simply not commercially available. We are at a crossroads for noninvasive coronary-artery imaging. It is widely anticipated that diagnostic use of radiographic catheter angiography will convert to noninvasive tests using MRI and CT. The research-and-development budgets for MRI versus CT at major health care companies will largely determine which test all of us will receive in coming years. Radiation will remain an issue, but this could potentially be avoided with MRI. If we do not like the options available, we could always try improvements in diet, more exercise, and weight loss. At least we have something of a choice.