Ernesto Castillo, MD.

Cardiac MR imaging has experienced significant technological advances over the past years. It currently can provide a comprehensive assessment of myocardial structure, function, perfusion, and viability with a capacity unmatched by any other single imaging modality. While MRI is an accepted and widely utilized tool for cardiovascular research, its clinical application has been limited but appears to be increasing given these recent advances. This article briefly reviews the expanding applications in this field with an emphasis on the evaluation of ischemic or acquired heart disease.

Approximately $1.7 billion was spent on all noninvasive cardiovascular imaging in Medicare recipients in 1998. 1 Echocardiography represented almost two thirds of all noninvasive cardiovascular imaging examinations while cardiac MR (CMR), introduced in the early 1980s, represented less than 0.1%. 1 Despite its limited use in clinical practice, CMR is attracting increased interest since some promising MR techniques recently became available outside of research centers. Here we review current MR applications for ischemic or acquired heart disease.


David A. Bluemke, MD, PhD.

The initial approach in most CMR applications is to obtain an accurate and artifact-free depiction of the cardiac chambers, myocardial wall, and valvular structures. It is also important to evaluate the pericardium and major adjacent vasculature, such as the pulmonary vessels and aorta.  Numerous techniques have been developed to depict cardiac morphology, which can generally be grouped by the appearance of the blood on the images as “black blood” and “bright blood” techniques.

ECG-gated spin-echo (SE) sequences have been for many years the main sequence used for evaluating cardiac morphology. Contrast between myocardium and blood is generated by a signal void of blood due to washout, hence the term black-blood images. Disadvantages of this technique are the limited temporal resolution and the sensitivity to respiratory and other motion-related artifacts due to the long scan times of several minutes. To overcome these problems, modified fast spin-echo (FSE) pulse sequences were introduced. A single-shot variation (SSFSE or HASTE), which reduces the scan time for an entire image to one heartbeat, allows  coverage of the entire heart within a single breath-hold. Another modification of the FSE sequence has been the addition of inversion-recovery (IR) pulses to null blood magnetization and to null fat, referred to as double or triple-IR FSE, respectively. These sequences are currently in widespread use in clinical CMR practice and provide an excellent, high-resolution depiction of cardiac anatomy, including such small structures as the valves (Figure 1). A promising, recent application of these techniques is the noninvasive imaging of the atherosclerotic plaque in the wall of the thoracic aorta and coronary arteries, as well as for coronary angiography. 2

Figure 1. Short-axis black blood double-IR FSE (left) and bright blood SSFP (right) images in a midventricular plane in a normal subject. Note the clear depiction of the trabeculation in the right ventricle and of the papillary muscles (at 2 o’clock and 7 o’clock) in the left ventricle.

Bright-blood imaging provides not only anatomic but also functional information, displayed in a cine format. These sequences include fast gradient-echo (GRE) and k-space segmented fast GRE, as well as steady-state free precession (SSFP) techniques. Bright blood in the GRE sequences depends on time-of-flight or inflow effects. Turbulent blood flow (ie, in areas of valvular stenosis or regurgitation) causes intravoxel dephasing, appearing as a dark jet. Also, areas of low blood velocity reduce the contrast between ventricular cavity and myocardium on gradient echo images. As a result, endocardial contour detection with automatic or semiautomatic segmentation algorithms is hampered. Recent improvements in gradient hardware technology allow SSFP cine images (Figures 1 and 2). Image contrast in SSFP is proportional to T2, and thus offers a higher intrinsic contrast between blood and myocardium than GRE techniques, even in patients with reduced flow. Acquisition times also are shorter compared to GRE techniques by a factor of two to three, which can be exploited to improve spatial and/or temporal resolution. To date, the most important application of the SSFP technique is for cine imaging of the heart because it depicts morphologic and functional abnormalities with greater clarity and provides greater diagnostic confidence than GRE techniques. 3


Figure 2. Oblique long-axis end-diastolic images obtained with bright blood cine 2D fast GRE (left) and SSFP (right) sequences within a single breath-hold. A stacked series of short-axis cine images covering the LV from apex to base can be used to measure accurately LV mass and end-diastolic and end-systolic volume as well as the resulting ejection fraction.

CMR imaging is a well-established technique for assessment of global and regional left ventricular (LV) function. CMR is regarded as the reference standard for assessing LV volumes, ejection fraction, and LV mass as it allows accurate and reproducible quantification. Despite its proven research utility, cardiac function evaluation by MR has not gained widespread use in clinical practice due to its higher cost and longer times for image acquisition and analysis compared to echocardiography. A potential solution may be provided by real-time techniques, which have advantages over the previously mentioned conventional cine MR by not requiring ECG-triggering, breath-holding, or navigator-echo gating. This reduces the setup time and duration of the examination, patient discomfort, and associated cost. 4 Ejection fraction is the most commonly used parameter of global assessment of LV systolic function in clinical practice for both diagnostic and therapeutic purposes. However, it does not take into consideration regional contractile dysfunction, which is commonly seen with ischemic heart disease and primary myocardial disease. Currently, regional function is usually measured with wall thickening obtained by echocardiography and is more precise than the subjective, visually estimated wall motion analysis. However, this parameter can be measured more accurately with the above-mentioned cine MR techniques. Wall thickening can also be considered a measurement of radial strain. Strain is defined as the percent fractional change in length between end-diastole and end-systole. Unlike echocardiography, which measures only the strain rate, strains can be directly obtained in CMR with use of myocardium tagging techniques (Figure 3). Tagged MRI allows intramyocardial, subendocardial, and subepicardial strain measurements in all three directions (circumferential, radial, and longitudinal) with respect to the heart’s geometry. Two-dimensional strain analysis is more accurate in describing segmental function than wall thickening analysis, with a sensitivity of 92% and 69% and specificity of 99% and 92%, respectively, in discriminating infarct-related dysfunctional myocardium from remote functional myocardium. 5 Despite its high accuracy, MR-tagging techniques are not frequently used in clinical practice for the assessment of coronary artery disease and myocardial viability. The reason for this is the time-consuming and labor-intensive postprocessing of the images, reducing its use to research purposes. One method recently proposed that overcomes this limitation is the harmonic phase (HARP) technique. Analysis of myocardial strains from tagged-MRI images based on this technique has been demonstrated to be fast (tenfold) and accurate, 6 but this method is still in development.


Figure 3. Short-axis GRE tagged images in a midventricular plane in the early (a) and end-systolic (b) phases. In MR tagging, planes of presaturation intersecting the myocardium are deposited prior to the imaging sequence, embedded in the tissue itself, and deform during systole (note the lack of deformation below the diaphragm and in the chest wall). Regional myocardial deformation can be tracked by how well the tag lines bend and come closer together during systole.

Compromised regional blood flow can be used as an indicator of possible reversible myocardial ischemia. CMR can perform myocardial perfusion studies with a higher resolution and lack of ionizing radiation compared to radionuclide techniques. The assessment of regional blood flow using MRI is performed with fast dynamic imaging of the first pass of contrast media (usually Gd-DTPA) through the myocardium. The state-of-the art techniques are hybrid sequences combining fast GRE and echo-planar imaging (EPI), which allow very short data acquisition times and adequate T1 contrast. Myocardium with the greatest concentration of contrast agent is assumed to be normally perfused and low signal areas are underperfused or ischemic. Under pharmacological stress of a vasoactive agent (dypiridamole or adenosine), the vascular resistance of a stenotic vessel is relatively high and results in a “vascular steal” phenomenon. Increased blood flow is redirected to the nonstenotic vessels, leading to an observable perfusion defect in the myocardial territory served by the stenotic vessel.

Initial methods of quantitative analysis have included the use of g-variate fits of signal upslope, measurements of mean transit times (MTT), and mean contrast enhancement (MCE). Although these techniques have been validated in animal models, more recent studies in human subjects reported that this approach is less than optimal. 7 One promising method is the use of a myocardial perfusion reserve index (MPRi) as a means to identify regions of reversible myocardial ischemia. Recent clinical studies using MPRi for detection of hemodynamically significant coronary disease found high diagnostic accuracy with sensitivities and specificities of 90% and 83%, respectively. 8 Depending on the acquisition technique chosen, qualitative examination of the MR perfusion data, and quantitative analysis of myocardial perfusion reserve, or both, can be used for clinical evaluation of coronary artery disease.


In the setting of myocardial infarction (MI) or ischemia, myocardium is considered viable if it is dysfunctional at baseline but recovers contractile function either spontaneously (stunned myocardium) or following revascularization (hibernating myocardium). Thus, revascularizing dysfunctional but viable myocardium leads to an increase of global left ventricular systolic function, reduces symptoms of congestive heart failure, and improves long-term prognosis. Identifying the presence and extent of viable myocardium allows a selection of those patients who would benefit most from revascularization strategies. Current techniques for viability include fluorine 18-labeled deoxyglucose positron emission tomography (FDG-PET), thallium 201 or technetium (99mTc-sestamibi) SPECT, and dobutamine stress echocardiography. They each have significant limitations, therefore, the recent interest in MRI techniques.

The simplest CMR technique involves the use of the above-mentioned cine sequences without contrast enhancement to cover the heart under resting conditions. The goal is to qualitatively assess or quantitatively measure end-diastolic wall thickness (EDWT) and systolic wall thickening (SWT). However, the marker EDWT is 92% sensitive but only 56% specific for predicting whether there will be functional improvement postrevascularization. 9 Better results (specificity up to 94% compared against PET criteria for viability) were obtained with dobutamine stress MRI. 10 The rationale of this technique is that viable myocardium shows augmented contractility in response to stimuli such as sympathomimetics, whereas necrotic or scarred tissue does not. Dobutamine is used in low dose (10 µg/kg/min) to detect contractile reserve (and hence viability) or high dose (up to 40 µg/kg/min) in order to detect ischemia indicating coronary artery disease. The major advantages of this technique over dobutamine echocardiography are the image quality and the possibility of quantifying the extent and severity of the wall motion abnormalities. However, limitations also exist with stress MRI, especially the subjectivity of the qualitative analysis. Transmural variation in mechanical function can be assessed with the addition of tagging techniques and may even further enhance the diagnostic accuracy of the technique. 

Figure 4. Perfusion and delayed-enhancement MRI in myocardial infarction (MI). The left short-axis first-pass perfusion image shows a subendocardial hypoenhanced area (arrowheads) in the left anterior descending artery territory, corresponding to myocardial obstruction within the MI. The right image reveals an extended area of non-transmural delayed hyperenhancement (solid arrows) indicating the extent of nonviable myocardium.

Contrast-enhanced MRI (CE-MRI) shows a regional heterogeneity of perfusion in human infarcts 11 (Figure 4, page 14). A few seconds (~20 s) after injection of a Gd-DTPA bolus, one sees within the infarct a region of hypoenhancement. Regions of hypoenhancement that persist for the first minute reflect the presence of microvascular obstruction (MO, ie, “no-reflow”), while the rest of the infarct increasingly enhances over time and becomes normoenhanced. Potential mechanisms for MO include post-reperfusion damage to the microcirculation at the arteriolar and capillary levels with secondary obstruction by erythrocytes, neutrophils, and necrotic debris causing inadequate tissue perfusion. At 10-20 minutes postcontrast, the infarct area becomes hyperenhanced (“delayed enhancement,” DE). DE after acute (1 and 3 days) and chronic MI (8 weeks) is an indicator of myocellular necrosis correlating precisely to the area of irreversible, nonviable tissue injury by histology (“bright is dead”). 12 The extent of both DE and MO has been shown to predict clinical outcome and prognosis in patients after acute MI. 13 It has also been recently shown that the CE pattern predicts contractile function of dysfunctional segments post-MI and provides details about the transmural extent of irreversibly damaged myocardium, predicting functional recovery postrevascularization. 14 A further advantage is the ability to visualize small, subendocardial areas of scar tissue that may be missed by nuclear techniques, including PET. 15 Myocardial viability can also be detected with the combined use of phosphorus (31P) and proton (1H) MR spectroscopy by measuring the concentrations of nonviable, MI-depleted metabolites phosphocreatine, ATP, and total and unphosphorylated creatine in myocardium. Thus, it allows assessment of in vivo myocardial creatine kinase metabolism and measurement of metabolite reductions seen with the loss of viability. 16 MR spectroscopy has been used mainly in research centers because of the limited availability of multifrequency MRI scanners.


Figure 5. Navigator-echo gated 2D image of the proximal coronary right coronary artery (RCA) in a healthy volunteer.

Coronary arteries are difficult to image noninvasively due to small size (2-4 mm in diameter), and tortuous and complex geometry, as well as continuous motion, except for a short period during mid diastole. During the last decade, numerous two-dimensional (2D) and three-dimensional (3D) coronary magnetic resonance angiography (CMRA) techniques have been developed and investigated (Figure 5). Breath-holding techniques without and with contrast enhancement have shown unsatisfactory results for detection of hemodynamically significant stenoses with sensitivities ranging from 63% to 90%. 17 Due to the long scan times, respiratory-gated acquisitions that enable imaging during tidal respiration were developed. The use of navigator-echo gating for 2D CMRA acquisitions appears promising as its short (2-3 min) examination times remove many of the respiratory gating and image quality concerns of longer navigator-gated 3D acquisitions. 18 Developments in navigator-gated 3D pulse sequences have made possible the first large multicenter trial of a CMRA technique. 19 Approximately 84% of the coronary arterial segments (proximal 3-5 cm of the main arteries) were sufficiently visualized and 83% of clinically significant (Ž50% lumen diameter) stenoses were correctly detected in comparison to catheter angiography. Technical issues including examination time, reliability, ease of use, and also education continue to be major practical concerns that have limited broad acceptance of these techniques.

Coronary MRA techniques are continuing to undergo development. Research concentrates on faster sequences (ie, modified SSFP techniques), improvements of signal to noise with higher magnetic fields (recently commercially available 3T instead of the current standard of 1.5 T), and improvement of signal intensity with newer intravascular blood-pool contrast agents. Because 20% to 40% of all diagnostic x-ray catheter angiograms reveal no clinically significant stenoses, the development of an accurate and robust CMRA technique combined with coronary artery wall imaging for plaque characterization would represent a significant improvement in the management of patients with suspected coronary artery disease.


As discussed above, there has been considerable technological improvement in CMR in the past several years. However, other competing imaging technologies are also evolving at a rapid rate. While it is unclear which modality will provide the best noninvasive coronary artery imaging, CMR clearly has many other useful applications that are proven and available today. Therefore, its incremental diagnostic value may replace other noninvasive cardiovascular modalities, especially echocardiography and nuclear medicine, for performing the evaluation of LV perfusion, viability, and global as well as regional function. The cost-effectiveness and impact on patient outcome still need to be proven for CMR to have a major role in the overall assessment of the patient with ischemic heart disease.

Ernesto Castillo, MD, is a research fellow.

David A. Bluemke, MD, PhD, is associate professor of radiology and clinical director of MRI at The Russell H. Morgan Department of Radiology and Radiological Sciences, The Johns Hopkins University School of Medicine, Baltimore.


  1. Levin DC, Parker L, Sunshine JH, Pentecost MJ. Cardiovascular imaging: who does it and how important is it to the practice of radiology? AJR Am J Roentgenol. 2002;178:303-306.
  2. 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;102:506-510.
  3. Pereles FS, Kapoor V, Carr JC, et al. Usefulness of segmented trueFISP cardiac pulse sequence in evaluation of congenital and acquired adult cardiac abnormalities. AJR Am J Roentgenol. 2001;177:1155-1160.
  4. Yang PC, Kerr AB, Liu AC, et al. New real-time interactive cardiac magnetic resonance imaging system complements echocardiography. J Am Coll Cardiol. 1998; 32:2049-2056.
  5. Götte MJW, van Rossum AC, Twisk JWR, et al. Quantification of regional contractile function after infarction: strain analysis superior to wall thickening analysis in discriminating infarct from remote myocardium. J Am Coll Cardiol. 2001;37:808-817.
  6. Garot J, Bluemke DA, Osman NF, et al. Fast determination of regional myocardial strain fields from tagged cardiac images using harmonic phase MRI. Circulation. 2000;101:981-988.
  7. Keijer JT, van Rossum AC, van Eenige MJ, et al. Magnetic resonance imaging of regional myocardial perfusion in patients with single-vessel coronary artery disease: quantitative comparison with (201) Thallium-SPECT and coronary angiography. J Magn Reson Imaging. 2000;11:607-615.
  8. Al-Saadi N, Nagel E, Gross M, et al. Noninvasive detection of myocardial ischemia from perfusion reserve based on cardiovascular magnetic resonance. Circulation. 2000;101:1379-1383.
  9. 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.
  10. 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;99:763-760.
  11. Lima JA, Judd RM, Bazille A, Schulman SP, Atalar E, Zerhouni EA. Regional heterogeneity of human myocardial infarcts demonstrated by contrast-enhanced MRI. Potential mechanisms. Circulation. 1995;92:1117-1125.
  12. Kim RJ, Fieno DS, Parrish TB, et al. Relationship of MRI delayed contrast enhancement to irreversibly injury, infarct age, and contractile function. Circulation. 1999;100:1992-2002.
  13. Wu KC, Zerhouni EA, Judd RM, et al. Prognostic significance of microvascular obstruction by magnetic resonance imaging in patients with acute myocardial infarction. Circulation. 1998;97:765-772.
  14. Kim RJ, Wu E, Rafael A, et al. The use of contrast-enhanced magnetic resonance imaging to identify reversible myocardial dysfunction. N Engl J Med. 2000;343:1445-1453.
  15. Klein C, Nekolla SG, Bengel FM, et al. Assessment of myocardial viability with contrast-enhanced magnetic resonance imaging: comparison with positron emission tomography. Circulation. 2002;105:162-167.
  16. Bottomley PA, Weiss RG. Noninvasive localized MR quantification of creatine kinase metabolites in normal and infarcted canine myocardium. Radiology. 2001;219:411-418.
  17. Li D, Zheng J, Bae KT, Woodard PK, Haacke EM. Contrast-enhanced magnetic resonance imaging of the coronary arteries. A review. Invest Radiol. 1998;33:578-586.
  18. Foo TK, Saranathan M, Hardy CJ, Ho VB. Coronary artery magnetic resonance imaging: a patient-tailored approach. Top Magn Reson Imaging. 2000;11:406-416.
  19. Kim WY, Danias PG, Stuber M, et al. Coronary magnetic resonance angiography for the detection of coronary stenoses. N Engl J Med. 2001;345:1863-1869.