Cardiovascular disease is the leading cause of death in the U.S. and most other industrialized countries.1,2 Cardiac catheterization and echocardiography are extensively used to study cardiovascular disease and represent a substantial portion of cardiac imaging-related expenditures in the U.S. Cardiac MRI, unlike x-ray angiography, is a noninvasive diagnostic imaging technique, with spatial and contrast resolution superior to echocardiography. It offers an alternative to most existing imaging techniques.
Cardiac MRI provides reliable anatomical and functional assessment of the heart and evaluation of myocardial viability and coronary artery disease.3 It also offers direct visualization and characterization of atherosclerotic plaques and diseased vessel walls and surrounding tissues,5-7 which is a hot topic in cardiovascular research.4 MRI can characterize the components of plaque and identify the presence of thrombus or calcium.
The use of MRI in cardiac applications has matured with the development of new techniques to control motion and flow artifacts. Ten years ago, most cardiac MRI suffered from respiratory motion compensation, but it is now possible to acquire quality images of the heart in a breath-hold. Alternatively, three-dimensional navigator free-breathing techniques offer robust respiratory motion compensation good enough to visualize coronary arteries.8-10 This has moved cardiac MRI out of the traditional cardiac gated spin-echo techniques into the newer segmented breath-hold or navigator techniques.
Improvements in MR hardware, such as fast high-duty cycle gradients and high-sensitivity radio-frequency coils, and the improved performance of gradient-intensive MR techniques make it possible in a clinical setting to evaluate cardiac morphology, function, and perfusion; myocardial viability information; and even coronary anatomy and flow. An integrated MRI approach to cardiac imaging can provide, in a noninvasive and time-efficient way, a wide array of relevant diagnostic information for patients with ischemic heart disease.
Despite the tremendous technological advances in hardware and imaging software, cardiac MR imaging remains underutilized in most clinical situations. This is mainly due to lack of exposure and experience on the part of clinicians and MR imagers. Major educational efforts are ongoing and will need to be expanded to make the modality more acceptable to clinicians. The American College of Radiology has initiated a series of cardiovascular imaging courses, offered four times a year (www.ACR.com).
ADVANCES IN TECHNOLOGY
Dedicated commercial cardiac MR scanners with a short bore, fast hardware, and gradient amplifier enhancements with shorter rise times (slew rate ~ 200 mtesla/m/msec), can perform superior cardiac MR studies. These gradient-intensive MR techniques allow higher speed imaging and high temporal resolution.
In addition, high-sensitivity RF coils and digital processing algorithms have been developed that speed image acquisition and reconstruction during cardiac MRI. New parallel data acquisition techniques, called SENSE and SMASH, use the phased-array multichannel coil technology to further improve spatial and temporal resolution.11,12
With SMASH (simultaneous acquisition of spatial harmonics) or SENSE (sensitivity encoding), several lines of data are acquired for each phase-encoding step, which is also referred to as a k-space trajectory. SMASH and SENSE differ from other techniques in which only one line of k-space data is acquired for each phase-encoding gradient step. SMASH imaging with a four-element array coil is four times faster and can be used to achieve almost real-time imaging. The maximum reduction in acquisition time is determined by the number of array coil elements. Thus, the heart can be scanned with higher temporal resolution and increased spatial resolution.
Real-time cine MR imaging with frame rates up to 40 images per second (acquisition time of 27 msec per image) has been reported using echo-planar imaging sequences with SMASH imaging.12 Hence, dynamic measurement of the transit of contrast agents with temporal resolution should make myocardial perfusion methods more precise. With all these advances, which allow freezing of heart motion during the cardiac cycle, cardiovascular MRI will become even easier and faster.
COMPETING MODALITIES
MRI is intrinsically 3-D and provides natural high contrast between the blood pool and cardiovascular structures without contrast media, on either black blood (spin-echo) or white blood (gradient-echo) imaging techniques. MRI of the heart and great vessels can be acquired during a breath-hold without contrast agents, providing exquisite visualization of the heart in patients with congenital heart disease or right ventricular dysplasia, and abnormalities of the pericardium, such as tumors and pericardial disease, or myocardium, both regional and global.
Cardiac anatomy is traditionally studied first with echocardiography. MRI offers a larger field-of-view than echocardiography, and does not have acoustic window access problems.
It also offers advantages over traditional CT, because MRI does not use ionizing radiation and can acquire high-resolution images in any plane. Disadvantages of CT include the ionizing radiation and its limitation to the axial plane, which allows only low-resolution sagittal and coronal views to be reconstructed. Newer ECG gated multislice CT scans also offer great potential for cardiac imaging.
MRI has superior spatial resolution to nuclear medicine.
CINE MRI
Assessment of ventricular function is central to the evaluation of patients with ischemic heart disease. The high performance of gradient-intensive MR techniques has made it possible to acquire short-axis cine MR imaging of the beating heart in real-time with high spatial and temporal resolution. Evaluation of cardiac function is thus possible with a series of short-axis cine images acquired from apex to the base of the heart to accurately measure left ventricular (LV) mass, end-diastolic volume (EDV), end-systolic volume (ESV), and ejection fraction (EF).
Several studies have documented that MR cardiac function analysis can replace echocardiography and cardiac nuclear medicine studies for the evaluation of wall motion and cardiac function. Traditionally, segmented breath-hold cine images required several heartbeats, with actual data acquisition lasting 100 msec per heartbeat. Advances in hardware and cardiac MR imaging techniques have made it possible to acquire a breath-hold cine image at the rate of 40 to 50 msec per frame and even faster. The 3-D gradient-echo cine MR image data set can provide global (EF) and regional ventricular function (wall thickening) parameters, as are provided by a cardiac nuclear medicine examination, but in less time and with higher spatial and temporal resolution.
Morphologic indices, assessment of systolic performance of the LV, LV mass, and cardiac function are strong predictors for cardiovascular disease morbidity or mortality. Cardiac MR imaging has been shown to accurately assess LV cardiac function.13 Its 3-D nature allows precise quantification of cardiac mass and volumes without geometric models or assumed formulas. This is an important improvement, as global LV function measurements using echocardiography or ventriculography depend on geometric assumptions that can be inaccurate. MR imaging is slowly becoming a gold standard for the evaluation of LV mass, volumes, and function in patients with ischemic heart disease.14
Another way to evaluate ventricular function is to compare flow measurement across the aortic and/or pulmonary outflow. This can be done with MR velocity mapping, which uses velocity-encoded imaging sequences. Phase-contrast cine MRI is routinely used to measure blood flow in great vessels and cardiac chambers, and this has enabled quantification of the function of both ventricles and the severity of valvular lesions. Phase-contrast cine MR is also used to study myocardial motion and deformation.
TAGGED MR IMAGES
Studies of myocardium motion have primarily used MR myocardial tagging techniques, in which a grid of "tag lines" is superimposed on the MR images, as described by Zerhouni15 in 1988. MRI myocardial tagging makes it possible to quantify the extent of regional heart wall motion abnormalities and thickening at rest and during stress.16 MR signal from the tissue in the tag planes is destroyed prior to the imaging sequence, and dark lines appear on the MR image where the tagging planes intersect the imaging plane. The dark tag lines persist during the cardiac cycle, deforming with the myocardium as it contracts during systole and expands during diastole. These tagged images can be used to track the motion of the myocardium.
The MR tags are noninvasive myocardial markers and must be regenerated at the onset of each contraction because they relax with the T1 of the heart, but persist beyond systole. The rate of loss of tag visibility is determined by the relaxation parameters of the myocardium tissue (T1 = 800 msec; T2 = 45 msec), and is a function of the imaging parameters used. The myocardial tags can persist up to 700 msec following the R wave, depending on the sequence used.
Tagging can be used with many types of imaging sequences, including spin-echo, gradient-echo, segmented gradient-echo with acquisition of multiple k-space lines per heart beat, and EPI sequences. Tag lines can be radial, which is ideal for the short-axis plane, or parallel for the long-axis plane. Tagging provides detailed information about regional cardiac function never before available with noninvasive techniques. Three-D maps of cardiac motion can be reconstructed over the entire left ventricle if tag displacement information in the three orthogonal tag directions is used to estimate 3-D myocardial displacement. MR tagging has been applied to evaluate regional ventricular mechanics early and late after myocardial infraction. Widespread use of MR tagging has been limited in part by the lack of good postprocessing tools, however.
BLACK BLOOD TECHNIQUES
MR black blood imaging techniques have been developed to improve segmentation of myocardium from the blood pool. These techniques decrease the signal from blood with reference to the myocardium and make it easier to perform cardiac chamber segmentation.
The black blood technique uses double-inversion pulses for magnetization preparation before the imaging sequence, which is turbo spin-echo. For dark blood preparation, a pair of nonselective and selective 180 degrees inversion pulses are used, followed by a long inversion time (TI) to null signal from inflowing blood. A second selective inversion pulse can also be applied with short inversion time to null the fat signal. These black blood techniques, referred to as double-inversion recovery turbo spin-echo or fast spin-echo, and double-inversion recovery STIR, are fully implemented on several commercial MR scanners.16
MYOCARDIAL ISCHEMIA
Dobutamine stress echocardiography is routinely used for evaluation of myocardial ischemia. These tests can also be performed with MRI.17,18 The most commonly used technique is breath-hold segmented k-space cine MRI. Using high-dose dobutamine in 208 patients who had coronary angiography, Nagel et al17 reported a sensitivity and specificity of 86.2% and 85.7% for MRI, compared with 74.3% and 69.8% for echocardiography. The authors concluded that the diagnostic accuracy for detection of wall motion abnormalities by dobutamine stress MRI is higher than that of dobutamine stress echocardiography. Tagged MR imaging of the myocardium can be very helpful for assessment of cardiac function during a stress test.
MR imaging of myocardial perfusion is an alternative to nuclear cardiology testing. The principle in both is to detect perfusion defects at rest and under stress. MRI, with its high resolution and ability to generate 3-D images, can track the distribution of a gadolinium contrast agent as it traverses the myocardium. Contrast-enhanced MR perfusion imaging can be reliably used for assessment of myocardial perfusion in patients with ischemic and infarcted myocardium. The uptake of contrast will be attenuated, in amplitude and rate, in regions of compromised flow.
Because exercise stress cannot be readily performed within the limited space of the MR magnet bore, most MR myocardial perfusion studies have been performed with vasodilators such as dipyridamole and adenosine. During most myocardial perfusion studies, the MR signal from the myocardium is first nulled to enhance the signal from the contrast media. To null the MR signal while maintaining good T1 weighting, several variants of spoiled gradient-echo perfusion sequences are used. One of more commonly employed methods involves an inversion pulse of 180 degrees and subsequent delay (TI ~ 300 msec) to null the myocardium signal before imaging with a spoiled gradient-echo sequence. Assessment of perfusion by this technique has correlated well with perfusion measured by microspheres.
In the past, adequate temporal resolution could not be achieved because the technique allowed only one slice to be imaged per breath-hold. It is now possible, however, to acquire one image per heartbeat at multiple slice locations with a T1 turboflash technique and also with T1-segmented EPI, allowing dynamic imaging of the bolus from right ventricle to left ventricle to myocardium and visualization of the coronary vessels.19,20 Acquisition of multiple slices per heartbeat allows tracking of the contrast bolus and provides the temporal resolution needed to characterize wash-in and wash-out kinetics.
Sensitivity and specificity of first-pass bolus MR imaging for the detection of perfusion deficits ranges from 74% to 92% and 87% to 96%, respectively, when compared with x-ray coronary angiography. By comparison, cardiac nuclear medicine scans show sensitivity and specificity of 65% to 82% and 75% to 81%, respectively.
MYOCARDIAL VIABILITY
Contrast-enhanced MRI has also been used to study patients with chronic and acute coronary artery disease (CAD), based on the observation that chronic infarcts hyperenhance on delayed images. In reperfused, or chronic, infarcts, myocardial signal intensity is increased or hyperenhanced.21 It has been shown that delayed hyperenhancement on MRI images of the myocardium three to 15 minutes after contrast injection correlates with nonviable myocardium areas in both acute and chronic infarcts (Figure 1).
CORONARY MRA
Coronary angiography (MRA) is an excellent tool for evaluating coronary artery anatomy but involves specific requirements because coronary vessels are small, tortuous in nature, often embedded in epicardial fat, and are associated with considerable physiologic motion. Good cardiac triggering is needed. Fat suppression should be uniform to increase the contrast between coronary arteries and the background.
Dedicated cardiac MR scanners with high-speed imaging techniques have come a long way in solving some of these problems. Optimal selection of imaging planes for coronary MR angiography is still an important issue for the newer targeted thin-slab 3-D breath-hold techniques (Figure 2)10 and also for the older segmented 2-D breath-hold acquisitions.23,24 Interactive real-time image plane selection techniques will have a dramatic effect on throughput of coronary MRA and MR examinations when they become commercially available. At that time, cardiac MR as a "one-stop" tool will become a clinical reality.
FLOW VELOCITY
Normal coronary circulation can increase its flow by a factor of four to six during stress. Volume flow and flow velocity in the coronary circulation increase with greater oxygen demands caused by exercise or in response to vasodilators and can be quantitated. This increase in flow is called the coronary flow reserve, which is the ratio of flow during maximum vasodilation to flow under resting conditions.25 Breath-hold phase-contrast cine MRI can be used to measure coronary blood flow.26 The measurements of coronary arterial flow and coronary flow reserve done noninvasively by MRI and invasively by ultrasound correlate well.
MR SPECTROSCOPY
Metabolic reserve can also be assessed by phosphorus MR spectroscopy (P-31 MRS), which offers another differentiating factor in the assessment of myocardial viability. The potential clinical applications of P-31 MRS for coronary artery disease is an active area of clinical research. Within seconds after reduction of oxygen supply, a decrease in phosphocreatine (PCr) and an increase in inorganic phosphate occur, the earliest metabolic responses in the myocardium.27 A decrease in PCr/ATP ratios can be detected in patients with reversible defects on thallium scintigraphy.
Measurement of metabolites such as creatinine by proton spectroscopy may also play a role in cardiac function. Scar tissue is characterized by reduced ATP content and also by almost complete loss of creatine. With further technical advances and improvement in resolution, it is possible that this biochemical stress test using MR spectroscopy can become the single tool for evaluating myocardial viability.
DR. DUERINCKX is chief of radiology and DR. KUMAR is the MRI physicist at the VA North Texas Healthcare System in Dallas.
---
References
1. Levin DC, Spettell CM, Rao VM, et al. Impact of MR imaging on nationwide health care costs and comparison with other imaging procedures. AJR 1998;170:557-560.
2. Shaaban AM, Duerinckx AJ. Magnetic resonance of coronary artery disease. Am J Geriatric Cardiol 1999;8:1-15.
3. Duerinckx AJ. Coronary MR angiography. Radiol Clin N Am 1999;37:273-318.
4. Fuster V, Ross R, Topol EJ. The pathogenesis of atherosclerosis. In: Fuster V, Ross R, Topol EJ, eds. Atherosclerosis and coronary artery disease. Lippincott-Raven 1996;chap 25:441-456.
5. Fuster V, Fayad, ZA, Worthley SG, et al. Magnetic resonance imaging and atherosclerotic disease (abstr). Proc Intl Soc Mag Reson Med 2000;8:139.
6. Fayad ZA, Fuster V. Characterization of atherosclerotic plaques by magnetic resonance imaging. Ann NY Acad Science 2000; 902:173-186.
7. Fayad ZA, Fuster V, Fallon JT, et al. Noninvasive in vivo human coronary artery lumen and wall imaging using black-blood magnetic resonance imaging. Circulation 2000;102(5):506-510.
8. Hartnell GG, Charlamb M, Cohen MC, et al. Breathhold cardiac MR imaging: improvements in image quality with fast SE, TurboSTIR, HASTE and breathhold cine MR angiography compared with standard techniques. Radiology 1996;201(P):273.
9. Bluemke DA, Boxerman JL, Atalar E, et al. Segmented k-space cine breath-hold cardiovascular MR imaging. Part 1. Principles and technique. AJR 1997;169(2):395-400.
10. Wielopolski PA, Geuns RJM, deFeyter PJ, et al. VCATS: Breath-hold coronary MR angiography with volume targeted imaging. Radiology 1998;209:209-219.
11. Weiger M, Pruessmann KP, Boesiger P. Cardiac real-time imaging using SENSE. Magn Reson Med 2000;43:177-184.
12. Sodickson DK, Griswold MA, Jakob PM. SMASH Imaging. MRI Clin N Am 1999; 7(2):237-254.
13. Lethimonnier F, Furber A, Balzer P, et al. Global left ventricular cardiac function: comparison between magnetic resonance imaging, radionuclide angiography, and contrast angiography. Invest Radiol 1999;34(3):199-203.
14. Kramer CM. Integrated approach to ischemic heart disease. Cardiol Clin 1998;16(2):267-276.
15. Zerhouni EA, Parish DM, Rogers WJ. Human heart: tagging with MR imaging-a method for noninvasive assessment of myocardial motion. Radiology 1988;169:59-63.
16. Simonetti OP, Finn JP, White RD, et al. Black blood T2-weighted inversion-recovery MR imaging of the heart. Radiology 1996;199:49-57.
17. Nagel E, Lehmkuhl HB, Bocksch W, et al. Noninvasive diagnosis of ischemia-induced wall motion abnormalities with the use of high-dose dobutamine stress MRI: comparison with dobutamine stress echocardiography. Circulation 1999; 99:763-770.
18. Baer FM, Theissen P, Schneider CA, et al. Dobutamine magnetic resonance imaging predicts contractile recovery of chronically dysfunctional myocardium after successful revascularization. J Am Coll Cardiol 1998;31:1040-1048.
19. Wilke NM, Jerosch-Herold M, Wang Y, et al. Myocardial perfusion reserve: assessment with multisection, quantitative, first pass MR imaging. Radiology 1997;204(2):373-384.
20. Finn JP. Recent developments in cardiac MRI. Intl Soc Mag Reson Med educational course syllabus 2000:140-147.
21. Duerinckx AJ. Myocardial viability using MR imaging: Is it ready for clinical use? AJR 2000;174:1741-1743.
22. Ramani K, Judd RM, Holly TA, et al. Contrast magnetic resonance imaging in the assessment of myocardial viability in patients with stable coronary artery disease and left ventricular dysfunction. Circulation 1998;98:2687-2694.
23. Duerinckx AJ. Coronary MR angiography. Radiol Clin N Am 1999;37(2):273-318.
24. Duerinckx AJ. Coronary arteries: How do I image them? In: Bogaert J, Duerinckx AJ, Rademakers F, eds. Magnetic resonance of the heart and great vessels-clinical applications. Berlin. Springer-Verlag, 1999:223-243.
25. Bogaert J, Dymarkowski S, Bosmans H. Myocardial perfusion. In: Bogaert J, Duerinckx AJ, Rademakers F, eds. Magnetic resonance of the heart and great vessels- clinical applications. Berlin, Springer-Verlag, 1999:133-151.
26. Hundley WG, Lange RA, Clark GD, et al. Assessment of coronary arterial flow and flow reserve in humans with magnetic resonance imaging. Circulation 1996;93(8):1502-1508.
27. Neubauer S. Myocardial metabolism. In: Bogaert J, Duerinckx AJ, Rademakers F, eds. Magnetic resonance of the heart and great vessels- clinical applications. Berlin, Springer-Verlag, 1999:155-
