This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by van der Wall, E. E. Articles by Bruschke, A. V.G. Search for Related Content PubMed PubMed Citation Articles by van der Wall, E. E. Articles by Bruschke, A. V.G.

(Circulation. 1995;92:2723-2739.)
© 1995 American Heart Association, Inc.

Articles

Magnetic Resonance Imaging in Coronary Artery Disease

Ernst E. van der Wall, MD; Hubert W. Vliegen, MD; Albert de Roos, MD; Albert V.G. Bruschke, MD

From the Departments of Cardiology (E.E.v.d.W., H.W.V., A.V.G.B.) and Radiology (A.d.R.), University Hospital Leiden, and the Interuniversity Cardiology Institute, Leiden, Netherlands.

Correspondence to Ernst E. van der Wall, MD, Department of Cardiology, Building 1, C5-P25, University Hospital Leiden, Rijnsburgerweg 10, 772333 AA Leiden, The Netherlands.

	Abstract

Top Abstract Introduction MR Imaging Techniques Coronary Artery Disease Conclusions References

 
Abstract The cardiovascular applications of nuclear magnetic resonance (MR) techniques in coronary artery disease have increased considerably in recent years. Technical advantages of MR imaging in comparison with other techniques are the excellent spatial resolution, the characterization of myocardial tissue, and the potential for three-dimensional imaging. This allows the accurate assessment of left ventricular mass and volume, the differentiation of infarcted tissue from normal myocardial tissue, and the determination of systolic wall thickening and regional wall motion abnormalities. Myocardial perfusion, metabolism, and inducible myocardial ischemia with the use of pharmacological stress also can be assessed by MR techniques. Future technical improvements in real-time imaging and development of noninvasive visualization of the coronary arteries and coronary artery bypasses will constitute a tremendous progress in clinical cardiology. Early detection and flow assessment of stenosed coronary arteries by MR angiography with the use of flow velocity measurements may outweigh the cost inherent to the MR imaging procedure. A particular strength of the MR technique is the potential to encompass cardiac anatomy, perfusion, function, metabolism, and coronary angiography in a single test. The replacement of multiple diagnostic tests with one MR test may have major effects on cardiovascular healthcare economics.

Key Words: magnetic resonance imaging • coronary disease • ischemia

	Introduction

Top Abstract Introduction MR Imaging Techniques Coronary Artery Disease Conclusions References

 
Magnetic resonance (MR) imaging is a unique noninvasive method for visualization of the heart.¹ The advantages of MR imaging in comparison with other imaging techniques are the clear delineation of the subendocardial and subepicardial margins of the cardiac walls, the characterization of myocardial tissue, the discrimination of intracardiac tumors and thrombi, and the direct visualization of pericardial structures. In particular, the amount of cardiac mass can be accurately measured, and diseases afflicting the myocardial walls are well defined by MR imaging. Technical advantages are the potential for three-dimensional imaging, the free choice of tomographic planes, and the lack of ionizing radiation. Disadvantages of MR imaging are the relatively long imaging times and the lack of obtaining bedside information. Furthermore, it is difficult to study critically ill patients, although patients with acute myocardial infarction have been safely studied within 24 hours after the acute event. Real-time MR imaging is not currently being used, but the rapid development of ultrafast imaging techniques may soon permit the application of the echo-planar techniques in clinical cardiology. MR imaging has opened new avenues for detecting cardiovascular abnormalities in an early stage of the disease process.² This review describes the value of MR imaging for detecting coronary artery disease. Although several other noninvasive imaging techniques such as echocardiography and radionuclide imaging are currently used on a routine basis in the assessment of patients with coronary artery disease, MR imaging may provide valuable information concerning the ischemic and infarcted heart, which is not available from other diagnostic techniques (Table 1).³ With respect to MR spectroscopy, the information gained on cardiac metabolism is unique, necessitating further exploration of this modality in the assessment of metabolic consequences of coronary artery disease.⁴ However, few clinical studies with MR spectroscopy in coronary artery disease have been performed, therefore this modality will be discussed briefly.

View this table: [in this window] [in a new window]   Table 1. Comparison of Technical Aspects of Noninvasive Imaging Methods

	MR Imaging Techniques

Top Abstract Introduction MR Imaging Techniques Coronary Artery Disease Conclusions References

 
Current MR imaging exists in two basic forms: spin-echo imaging and gradient-echo or cine MR imaging. While the spin-echo technique shows lack of signal intensity in vascular compartments with rapid blood flow ("black blood imaging"), the cine MR imaging technique depicts flowing blood as a bright signal ("white blood imaging"). Newer developments in MR imaging are ultrafast imaging techniques and myocardial tagging. For a better understanding of the theoretical background of the routinely used MR techniques, one is referred to publications that deal more specifically with the basic physics involved in MR imaging and spectroscopy.^{1 2 3 4 5 6 7 8 9} The clinical aspects of the MR techniques used in cardiology are discussed briefly.

Spin-Echo Imaging
Spin-echo imaging is the MR imaging technique of choice in assessing left ventricular mass by providing a three-dimensional, direct visualization of the myocardium with excellent mural edge discrimination (Fig 1). Cardiac and vascular anatomy have been examined in normal subjects and in patients with various forms of cardiovascular disease. Muscle mass and chamber dimensions correlate well with those obtained with echocardiography and contrast ventriculography.^{10 11 12 13 14 15 16 17 18 19} Left and right ventricular volumes, left ventricular ejection fractions and regional left ventricular function can accurately be measured.^{10 11 12 13 14 15 16 17 18 19 20 21 22 23 24} In addition, wall thickness and thickening, as measures of myocardial viability, may be determined with the use of MR spin-echo imaging.^{25 26 27} At present, MR spin-echo imaging is predominantly used for the evaluation of static phenomena such as morphological appearance, myocardial tissue characterization (with and without paramagnetic contrast agents), assessment of wall thickness and left ventricular mass, and left ventricular thrombus, whereas cine MR imaging is the technique of choice for evaluation of cardiac function (Fig 2).

View larger version (152K): [in this window] [in a new window]   Figure 1. Orthogonal cardiac short-axis spin-echo magnetic resonance image of a patient without cardiac disease. Myocardial walls of the left and right ventricle are clearly depicted.

View larger version (175K): [in this window] [in a new window]   Figure 2. Transverse cardiac spin-echo magnetic resonance image of a patient with an anterior wall infarction with apical and septal involvement. For this T2-weighted image, a multi-echo study (TE 30-60-90-120 ms) was performed. Note the aneurysmal dilatation of the apicoseptal wall with increased signal intensity on the images with echo times 60 ms (upper right), 90 ms (lower left), and 120 ms (lower right).

Myocardial Tissue Characterization
Because of its capacities for tissue characterization, MR imaging is well suited for the evaluation of myocardial ischemia and infarction. Myocardial ischemia and infarction are associated with increased myocardial signal as a result of prolonged tissue relaxation times T1 and T2 (T1, longitudinal relaxation time; T2, transverse relaxation time).²⁸ In several studies, correlations were observed between increases in T1 and T2 and the increase in tissue water content.^{29 30} Although it is clear that tissue water content plays a major role in T1 and T2 changes, it has been suggested that other factors also may be important such as changes in free radicals or changes in magnetic susceptibility.^{31 32 33}

Contrast Agents
The quality of MR imaging has been greatly improved by the use of paramagnetic contrast agents.³⁴ Paramagnetic compounds cause a shortening of both the T1 and T2 relaxation times, with the reduction in T1 relaxation time predominating. The magnitude of the change in relaxation time is influenced by the magnetic field strength, the paramagnetic agent used, and the agent concentration. Local tissue perfusion is thus delineated in a manner analogous to standard indicator dilution methods, with more pronounced effects seen in areas of highest contrast agent concentration. Despite the ability to generate images with varying image contrast, when the relaxation parameters T1 and T2 are used, the detection of acute ischemia with unenhanced MR imaging does not occur until several hours after coronary occlusion. Therefore, paramagnetic contrast agents have been used to define functional and perfusion abnormalities in the early stage of acute myocardial ischemia and infarction.^{35 36 37 38 39 40 41 42 43 44 45} The first studies of contrast-enhanced MR imaging used manganese chloride to evaluate ischemic and infarcted myocardial zones in canine hearts.^{46 47} At present, gadolinium-containing contrast agents (labeled with DTPA, DOTA, BOPTA, or albumin) are being used widely.^{48 49 50 51 52 53 54 55 56 57}

Cine MR Imaging
With cine MR imaging the normal blood pool shows high signal intensity in all phases of the cardiac cycle and provides a constant contrast with the less intense myocardium, unlike spin-echo imaging in which intraluminal signal is usually low and may be variable and inseparable from adjacent myocardial tissue because of flow artifacts. Visual evaluation of global and regional wall motion is facilitated by viewing the cinematic display of the MR images. Basically, the cine MR imaging technique may be more accurate for defining regional myocardial (dys)function than contrast angiography, since the latter depends upon the evaluation of wall motion only. Cine MR imaging appears to be the MR technique of choice for assessing left ventricular function.^{58 59} Cardiac chamber volumes, ejection fraction, and regional wall motion and thickening can be reliably determined by cine MR imaging, with high reproducibility.⁶⁰ All parameters correlated well with two-dimensional echocardiography and contrast ventriculography.^{61 62 63 64} Pattynama et al^{65 66} showed that cine MR imaging allowed the accurate assessment of right ventricular volumes and right ventricular mass.

Ultrafast MR Imaging
Ultrafast imaging refers to a group of techniques developed with the main purpose of acquiring images very rapidly. Because much of the difficulty in cardiac MR imaging arises from respiratory motion, imaging fast enough to be completed in a single breathhold reduces this problem. Ultrafast gradient-echo pulse sequences have been used recently for cardiac application and can be referred to as turbo fast gradient-echo imaging.^{67 68 69 70 71} In particular, the use of commercially available rapid imaging sequences with fast low-angle shot imaging has enabled more widespread study of myocardial perfusion and wall motion with MR imaging.^{68 69} During fast gradient-echo acquisitions, one can simulate cardiac gating by reordering and segmentation of the MR image data acquisition (so-called segmented k-space acquisition). This allows for almost total elimination of motion by limiting the acquisition to 50 to 120 ms at end diastole.⁷⁰ The fast gradient-echo techniques also allow the acquisition of end-systolic and end-diastolic phases of the cardiac cycle to calculate cardiac volumes and mass.⁷¹ Other ultrafast MR imaging techniques, such as echo-planar MR imaging, which obtain images in approximately 50 ms (one-shot imaging), are also available.⁷² Within a single breathhold, the entire heart at a specified time in the cardiac cycle may be covered in approximately half a minute.⁷³ Coupled with the bolus administration of contrast media and the acquisition of first-pass images of the left ventricular myocardium, ultrafast MR imaging has great potential in the assessment of regional myocardial wall motion and perfusion (Fig 3). It enables the study of fast physiological processes, such as cardiac first-pass effects. The limitations of the current technique are its inability to quantitate myocardial flow and the acquisition of images at only a limited number of cardiac levels (currently maximum 3) during a single bolus injection. Future technical improvements such as echo-planar imaging should allow for more tomographic sections to be acquired at several levels with each heartbeat during the first pass of a contrast agent.⁷⁴

View larger version (129K): [in this window] [in a new window]   Figure 3. Series of six sequential ultrafast magnetic resonance images showing signal enhancement after administration of Gd-DTPA: Baseline image (upper left), right ventricular cavity (upper middle and upper right), pulmonary vasculature, left ventricular cavity and aorta (lower left and lower middle), and myocardium (lower right). Courtesy F.P. Van Rugge.

Myocardial Tagging
Recently, MR imaging methods have been described that permit the determination of the absolute motion and thickening of specific myocardial segments. These involve the use of myocardial tagging as first described by Zerhouni et al⁷⁵ and Bolster et al⁷⁶ and subsequently enhanced by Axel and Dougherty.^{77 78} Myocardial tagging involves localized radiofrequency saturation of the myocardial tissue before acquiring images, which allows the monitoring of the progressive distortion of the myocardium during the course of the cardiac cycle. The earlier tagging sequences (eg, DANTE⁷⁹ ) have been modified to increase the number of saturation bands, narrow the bands, and improve their persistence throughout the cardiac cycle. The most frequently used tagging sequence is called spatial modulation of magnetization (SPAMM),^{77 78} which produces images with a regular pattern of stripes that move with the cardiac wall during the cardiac cycle. It provides a unique method for analyzing regional ventricular strain and the quantitation of regional myocardial function such as the absolute motion and thickening of specific myocardial segments.⁸⁰ This can be used to evaluate myocardial rotational deformation, ventricular nonuniformity, and differences in subendocardial and epicardial wall motion, both in normal and ischemic myocardium.^{81 82 83} Based on these findings, it is likely that myocardial tagging will become a reference standard for assessment of wall motion and wall thickening in the setting of acute myocardial ischemia and infarction. Clinical studies in patients with coronary artery disease are awaited.

MR Spectroscopy
MR spectroscopy is an exciting tool for evaluation of cardiac metabolism by direct measurement of ischemia-induced changes of high-energy phosphates and the intracellular pH in in vivo animal models with the use of surface coils directly applied to the surface of the heart.⁸⁴ During ischemia, adenosine triphosphate (ATP) and phosphocreatine (PCr) levels decrease, whereas the level of inorganic phosphate increases.⁸⁵ After brief periods of ischemia, PCr levels recover during reperfusion.⁸⁶ After the onset of ischemia, the concentration of PCr decreases much faster than that of ATP, resulting in a rapid increase of the PCr/ATP ratio.⁸⁷ At present, only a few spectroscopy data are available that apply these experimental data to patients with coronary artery disease.^{88 89} With the use of isometric stress testing^{88 89} a significant decrease in PCr/ATP ratio was obversed in ischemic myocardial regions. Quantification of metabolism in humans may be difficult because volumes of interest are relatively large compared with myocardial wall thickness. As a result, there are no reliable data available in patients with a previous myocardial infarction who may have considerable wall thinning, which precludes reliable acquisition of MR spectra. At present, clinical applications of cardiac MR spectroscopy in patients with coronary artery disease are sparse, and its value remains to be proven in larger studies.⁹⁰

	Coronary Artery Disease

Top Abstract Introduction MR Imaging Techniques Coronary Artery Disease Conclusions References

 
The use of MR imaging in coronary artery disease falls into four main categories: (1) evaluation of acute myocardial ischemia and infarction, (2) assessment of the sequelae of myocardial infarction, (3) evaluation of coronary artery bypass grafts, and (4) visualization of the coronary arteries.

Acute Myocardial Ischemia and Infarction
Experimental Studies in Acute Myocardial Ischemia
MR imaging allows the assessment of infarct size based on different T2 relaxation times between infarcted and normal tissue.^{91 92 93 94} Serial MR imaging of left ventricular infarct size 3 and 21 days after coronary artery ligation with the use of T2 measurements correlated well with histopathologically assessed infarct size.⁹⁵ MR imaging of dogs with reperfused myocardium showed a significant increase in signal intensity and T2 relaxation times as early as 30 minutes after reperfusion,^{96 97 98 99 100 101 102} indicating that MR imaging detects ischemic myocardial areas soon after coronary occlusion, thereby providing a method to discern reperfused viable myocardium. Based on these experimental findings, a T2 strategy has been advocated to evaluate healing patterns in patients following reperfusion after thrombolytic therapy.¹⁰³

The paramagnetic contrast agent Gd-DTPA has been shown to enhance contrast of ischemic and infarcted myocardium in animal experiments.^{41 43 46 47} The contrast enhancement in the ischemic area probably is caused by differences in wash-in and wash-out of Gd-DTPA from normal and ischemic myocardium. In acutely damaged myocardium, the increased accumulation of Gd-DTPA may be related to one or more of the following factors: decreased blood flow, increased tissue blood volume, enhanced size of the extracellular space, and increased permeability of the capillaries, all of which cause slow wash-out from the infarcted zone. By 10 to 15 minutes after Gd-DTPA injection, it has largely washed out of the normal myocardium, whereas it remains in the infarcted zone, suggesting that MR imaging in acute myocardial infarction should be performed more than 15 minutes after administration of Gd-DTPA. Gd-DTPA remains outside the cells and is excreted by glomerular filtration.³⁶ Gd-DTPA has been studied in several experimental models of myocardial ischemia that primarily differ from each other in the duration of coronary artery ligation, the time period between contrast administration and imaging, and the presence or absence of reperfusion.^{49 50 51 52 53} These experimental studies with Gd-DTPA demonstrated that changes in relaxation times occur very early (2 minutes) after coronary artery occlusion, implying that Gd-DTPA allows the detection of early myocardial ischemia even before the onset of myocardial edema formation or the development of irreversible damage. These studies also suggest that Gd-DTPA may be useful to outline distribution of regional myocardial blood flow. In a study by Miller et al⁵⁴ using Gd-DTPA–enhanced MR imaging, it was possible to measure myocardial flow reserve during pharmacological dilatation by dipyridamole. There was a significant correlation between changes in Gd-DTPA–enhanced MR signal and microsphere-determined myocardial blood flow. Further experimental studies have shown that the use of Gd-DTPA may discriminate between occlusive and reperfused infarcts, based on differences in signal intensities.^{55 56 57} Moreover, administration of Gd-DTPA early after reperfusion allowed the identification of the area at risk by selective concentration of Gd-DTPA in reperfused myocardium.^{104 105} Holman et al,¹⁰⁶ in 21 isolated rat hearts, compared distribution of Evans blue staining and Gd-DTPA–induced contrast enhancement in ischemic and reperfused myocardium and showed the excellent capability of Gd-DTPA to identify ischemia and reperfusion by contrast enhancement. Nishimura et al¹⁰⁷ measured infarct size in canine hearts both by Gd-DTPA–enhanced MR imaging and indium-111 labeled antimyosin. Gd-DTPA showed significant contrast enhancement of the infarcted area, and the extent of the contrast enhancement expressed infarct size precisely. Van Dijkman et al¹⁰⁸ showed that Gd-DTPA–enhanced MR imaging identified infarcted myocardium with great sensitivity in an in vivo porcine model.

In summary, experimental studies using spin-echo MR imaging may identify both reperfused and nonreperfused myocardium by its tissue characterization capabilities. Gd-DTPA–enhanced MR imaging shows improved contrast enhancement of ischemic and infarcted myocardium, especially on T1-weighted images. It allows the distinction between normal and ischemic myocardial tissue and can be used to delineate the infarcted area and to calculate infarct size.

Clinical Studies in Myocardial Infarction
Clinical studies in patients with documented myocardial infarction have shown T1 and T2 alterations in infarcted myocardium.^{109 110 111 112 113 114 115 116 117 118} Johnston et al¹¹⁰ studied 34 patients 3 to 30 days after myocardial infarction and showed that regional increase of signal intensity was consistent with the electrocardiographic location of the infarction and with the presence of hypokinetic segments on the left ventriculogram. Fisher et al¹¹¹ showed in 29 patients 3 to 17 days after myocardial infarction prolonged T2 relaxation times in infarcted myocardial regions. On the other hand, they observed that increased signal intensity on T2-weighted images may be very difficult to distinguish from slowly moving intraventricular blood flow. Ahmad et al¹¹² showed that T2 prolongation might not be a specific marker for acute myocardial infarction and can also be observed in abnormally perfused myocardial segments of patients with unstable angina. Been et al^{113 114} showed in 41 patients with acute myocardial infarction that maximum T1 values were observed at 2 weeks after the acute onset, suggesting that the increase of T1 reflects cellular infiltration as much as or more than tissue edema. No differences in T1 values were observed between the patients with or without reperfusion, indicating that alterations of T1 are complex and may bear no relationship to specific histological findings. In the absence of any histological confirmation, these statements remain purely speculative. Studies by Postema et al¹¹⁵ and Krauss et al^{116 117} showed in patients with acute myocardial infarction, who underwent MR imaging studies with a mean of 8 days after the acute event, that regional T2 abnormalities in 82% of patients correlated with the presence, location, and extent of thallium-201 perfusion defects at rest. In a subsequent study by Krauss et al,¹¹⁸ good agreements were found between enzymatic infarct size, thallium-201 scintigraphy, radionuclide angiography, and MR findings.

In addition to changes in T1 and T2 relaxation times as indices for tissue characterization, other characteristics can be used to indicate infarcted myocardial areas, such as increased signal intensity, ventricular cavitary signal, and regional wall thinning. Filipchuk et al¹¹⁹ showed increased myocardial signal intensity in 88%, cavitary signal in 74%, and regional wall thinning in 67% of 27 patients with acute myocardial infarction. Of the three features, wall thinning was the most predictive of and specific for acute myocardial infarction (Fig 4). White et al^{120 121} showed in patients with recent myocardial infarction a good correlation between MR imaging and two-dimensional echocardiography for demonstrating regional wall motion abnormalities; the extent of regional wall thinning by MR imaging could be used to measure infarct size. Wisenberg et al¹²² showed in 66 patients 3 weeks after acute infarction that infarct size could very well be determined by MR imaging based on signal intensity. They demonstrated that in the 41 patients who had received acute streptokinase therapy, a significant reduction in MR-measured infarct size was observed compared with the patients without thrombolytic therapy. Johns et al¹²³ assessed MR infarct size in 20 patients based on signal intensity at a mean of 9 days after the acute onset of symptoms. MR infarct size correlated very well with the extent of the region with severe hypokinesia visualized by left ventricular angiography. Turnbull et al¹²⁴ compared MR imaging, based on T1 maps, with enzymatic infarct size, technetium-99m pyrophosphate scintigraphy, and radionuclide angiography in patients 5 to 7 days after myocardial infarction. The authors found a good agreement between infarct size detected by MR imaging and that assessed by the radionuclide techniques.

View larger version (149K): [in this window] [in a new window]   Figure 4. Short-axis cardiac spin-echo magnetic resonance image of a patient with a 2-day-old myocardial infarction of the inferoposterior wall. There is marked wall thinning of the inferoposterior wall and a clear dilatation of the left ventricle.

Acute myocardial infarction also has been studied with the use of contrast-enhanced MR imaging. Most clinical experience has been obtained with Gd-DTPA, which can be safely used in patients with coronary artery disease and generally provides a better image quality than unenhanced T2-weighted images.⁴⁸ Eichstaedt et al¹²⁵ were the first to show, in 26 patients with subacute myocardial infarction, that the 11 patients who were studied with Gd-DTPA 5 to 10 days after the acute event (at 10, 25, and 45 minutes after administration of Gd-DTPA) had a 70% average increase of signal intensity within zones of infarcted myocardium, while only a 20% increase of signal intensity in normal myocardial tissue was observed. The other 15 patients were imaged later in the course of infarction and did not show differences in intensity ratio between infarcted and normal tissue. These findings were corroborated by Nishimura et al,¹²⁶ who studied infarct patients in the subacute phase with MR imaging and Gd-DTPA 5 to 10 minutes after administration, at an average of 5, 12, 30, and 90 days after the acute event. Increased signal intensity in the infarcted area was observed at 5 and 12 days, implying that only subacute myocardial infarcts show significant accumulation of Gd-DTPA. In our institution, preliminary studies^{127 128} showed that the signal intensity of infarcted versus normal myocardium was significantly greater after Gd-DTPA administration than before Gd-DTPA both by visual and computer-assessed analysis. Maximal contrast was observed at 20 to 25 minutes after administration of Gd-DTPA (Fig 5). Van Dijkman et al¹²⁹ also showed that signal intensity of Gd-DTPA was significantly increased in the infarcted areas of patients who were studied more than 72 hours after the acute onset, indicating increased accumulation of Gd-DTPA in a more advanced stage of the disease process. In a subsequent study by Van Dijkman et al¹³⁰ in 84 patients with acute myocardial infarction, it was shown that Gd-DTPA enhancement improved visualization of infarcted areas up to 6 weeks after onset of symptoms and had a maximal effect within 1 week after infarction. Holman et al¹³¹ showed a good correlation between infarct size measured with gadolinium-enhanced MR imaging and enzymatically determined infarct size in 24 patients 3 to 7 days after the acute event (Fig 6).

View larger version (113K): [in this window] [in a new window]   Figure 5. A, Transverse cardiac spin-echo magnetic resonance (MR) image of patient with anteroseptal wall infarction before administration of Gd-DTPA. Apart from some apical thinning, no clear MR signs of myocardial infarction are recognized. B, Same MR image 20 to 25 minutes after administration of Gd-DTPA. Contrast enhancement is visible in the anteroseptal area with extension to the apex.

View larger version (86K): [in this window] [in a new window]   Figure 6. A, Computer-constructed contours of subepicardial (1) and subendocardial (2) borders on cardiac spin-echo magnetic resonance (MR) image after administration of Gd-DTPA in patient with anterior wall infarction. B, After subtraction of mean cardiac signal intensity (+2 SD), the MR image shows marked contrast enhancement of Gd-DTPA in anteroapical area (3). C, Summing up the extent of contrast enhancement in the different tomographic slices (anterior view) covering the complete left ventricle, an estimate of infarct size can be obtained. D, Correlation between infarct size determined by MR imaging (MRI) and enzymatic infarct size calculated from cumulative release of -hydroxybutyrate dehydrogenase (HBDH) activity in plasma. Regression line and 95% confidence interval are indicated. Note that line of identity (dashed line) is within the 95% confidence interval. (From E.R. Holman et al. Am J Cardiol. 1993;71:1036-1040. Reproduced with permission.)

These encouraging results have led to initiation of clinical studies to determine whether the use of Gd-DTPA allows the discrimination of reperfused versus nonreperfused myocardial areas. In an initial report by Van der Wall et al¹³² using Gd-DTPA–enhanced MR imaging in 27 patients after thrombolytic therapy for acute myocardial infarction, it was shown that signal intensities measured 25 minutes after Gd-DTPA administration did not differ between reperfused and nonreperfused myocardial areas. Van Rossum et al¹³³ studied patients with acute myocardial infarction after thrombolytic therapy, and they measured intensity ratios 6 to 8 minutes after injection of Gd-DTPA. They did observe a significant difference in signal intensity ratios between infarcted regions subtended by occluded coronary arteries and reperfused vessel regions, indicating that assessment of the early dynamics of contrast enhancement using Gd-DTPA MR imaging may identify successful reperfusion. In a study by De Roos et al¹³⁴ it was observed that the morphological appearance of contrast enhancement by Gd-DTPA may provide some clues as to the presence or absence of reperfusion; reperfusion goes along with a homogeneous aspect, while lack of reperfusion may be visualized as a heterogeneous enhancement of contrast. Apart from these morphological characteristics, De Roos et al¹³⁵ used MR imaging with Gd-DTPA to show that infarct size was significantly smaller in patients with documented reperfusion than in patients without reperfusion (Fig 7).

View larger version (27K): [in this window] [in a new window]   Figure 7. Bar graph: After thrombolysis, a significant reduction of magnetic resonance imaging (MRI)–determined infarct size is observed both in the early phase (3 days) and in the late phase (5 weeks) after the acute event. Infarct size was calculated from the percent left ventricular (LV) myocardium that showed contrast enhancement after Gd-DTPA administration.

In summary, acute myocardial infarction is associated with prolonged T1 and T2 relaxation times. In addition to increased relaxation times, other morphological MR imaging features are valuable, of which regional myocardial wall thinning is most specific. The use of paramagnetic contrast agents has considerably improved the detection of myocardial infarction. Particularly in patients within 1 week after myocardial infarction, the effect of thrombolytic therapy can be assessed with the use of the early dynamics of contrast enhancement after administration of Gd-DTPA and by the accurate determination of infarct size. Disadvantages include the difficulty of monitoring acutely ill patients, lack of ability to do bedside studies, and contraindications for patients with pacemakers.

Assessment of Myocardial Function
The cine or gradient-echo MR imaging technique can be used for detection of myocardial ischemia by analysis of global and regional cardiac function. Abnormal wall motion and more specifically abnormal wall thickening indicate diminished regional myocardial function.¹³⁶ In a study by Pflugfelder et al,¹³⁷ 13 normal subjects and 15 patients with coronary artery disease were studied by cine MR imaging to document and quantitate regional left ventricular wall motion abnormalities. Abnormal wall motion was observed in 40 of 90 segments in patients with coronary artery disease, which correlated well with results of echocardiography or contrast ventriculography. The overall systolic wall thickening in the normal subjects was 48±28%, in the normal segments of the patients 43±31%, in hypokinetic zones 6±18%, in akinetic zones -4±24%, and in dyskinetic zones -13±25%. Thus, the absence of systolic wall thickening proved to be a very specific marker of regional myocardial dysfunction. Lotan et al¹³⁸ studied 59 patients with suspected coronary artery disease with both biplane cine MR imaging and biplane cineangiography. In the right anterior oblique view, there was agreement in 96% of 275 segments and in the left anterior oblique view in 92% of segments. Meese et al¹³⁹ studied 25 patients within 7 days after acute myocardial infarction using cine MR imaging, left ventricular contrast angiography, and radionuclide angiography. They showed that left ventricular ejection fraction by cine MR imaging correlated better with the ejection fraction by left ventriculography (r=.94) than ejection fraction by radionuclide angiography (r=.82). Compared with left ventriculography, the concordance in regional wall motion was similar for both cine MR imaging (69%) and radionuclide angiography (65%).

The capability of cine MR imaging to provide functional information about the state of pathologically altered myocardium in combination with assessment of diastolic wall thickness and systolic wall thickening makes it suitable for identification of myocardial viability.^{140 141 142} Data from Baer et al,^{143 144} who compared wall thickness measurements by cine MR imaging with technetium-99m methoxy-isobutyl-isonitrile (MIBI) tomographic imaging, showed a high concordance in patients with large chronic Q-wave infarcts, particularly in patients with anterior wall infarcts. In a recent study, Baer et al¹⁴⁵ compared low-dose dobutamine MR imaging with positron emission tomography in 35 patients with myocardial infarction (>4 months old). They showed that MR imaging was very accurate in assessing myocardial viability. These findings are supported by Perrone-Filardi et al,^{146 147} who used positron emission tomography with fluorine–18-fluorodeoxyglucose. In most regions with reduced end-diastolic wall thickness and absent wall thickening, absence of metabolic activity was shown, indicating the suitability of MR imaging in the evaluation of myocardial viability.

In recent years, pharmacological stress has been applied in MR imaging for detection of functional abnormalities in patients with coronary artery disease, since physical exercise during MR imaging is difficult because of motion artifact and space restriction. Compared with dipyridamole as a vasodilating agent for producing perfusion abnormalities, dobutamine appears to be a more appropriate agent for eliciting wall motion abnormalities.¹⁴⁸ Pennell et al¹⁴⁹ studied 22 patients with coronary artery disease both by dobutamine cine MR imaging and thallium tomography. Comparison of perfusion defects and wall motion abnormalities during stress showed 90% agreement, and dobutamine infusion was well tolerated in all patients.

Van Rugge et al,¹⁵⁰ in 23 healthy volunteers, identified wall motion dynamics and provided calculations of segmental wall thickening and hemodynamic parameters by using dobutamine stress imaging. In 37 patients with coronary artery disease, Van Rugge et al¹⁵¹ showed an overall sensitivity of 81% and a specificity of 100% when using dobutamine MR imaging; in patients with single-, two-, and three-vessel disease the sensitivity values were 75%, 80%, and 100%, respectively. In a subsequent study in 39 consecutive patients with clinically suspected coronary artery disease referred for coronary arteriography and in 10 normal volunteers, it was shown that dobutamine cine MR imaging identified wall motion abnormalities by quantitative analysis using the centerline method¹⁵² ; the sensitivity, specificity, and accuracy were 91%, 80%, and 90%, respectively. These findings were corroborated by Baer et al,¹⁵³ who studied 28 patients with dobutamine cine MR imaging and found an overall sensitivity of 87% and a specificity of 100% for the detection of coronary artery disease. In a recent study, Baer et al¹⁵⁴ compared the findings of dobutamine MR imaging with findings of dobutamine with technetium-99m MIBI tomographic imaging in 35 patients with coronary artery disease; a high concordance between the two imaging modalities was found with respect to the detection of a dobutamine ischemic response. These studies illustrate the feasibility of cine MR imaging to perform stress imaging and to detect the functional sequelae of reversible myocardial ischemia.

In summary, determination of wall motion and wall thickening by cardiac cine MR imaging may play an important role in the accurate detection and functional characterization of patients with suspected or known coronary artery disease. Cine MR imaging with the use of pharmacological stress may constitute a new modality to detect coronary artery disease. With the use of centerline analysis, accurate quantitative information can be obtained from regions that show reduced wall thickness and thickening caused by stress-induced myocardial ischemia. Although MR imaging–determined reduction in end-diastolic wall thickness and absence of wall thickening may identify most viable segments, it is currently not the optimum approach, and metabolic information obtained by the use of MR spectroscopic methods may improve the specificity of functional analyses alone. Furthermore, more advanced technical developments are required before pharmacological stress cine MR imaging becomes a serious challenge to pharmacological stress radionuclide perfusion imaging or to two-dimensional echocardiography.

Assessment of Myocardial Perfusion
Most MR imaging studies on myocardial perfusion have been performed with the use of ultrafast MR imaging. Atkinson et al¹⁵⁵ demonstrated that a T1-weighted ultrafast MR imaging technique can provide adequate temporal and spatial resolution to permit first-pass perfusion studies of the heart. In an isolated perfused rat heart model with Gd-DTPA as contrast agent, marked differences in contrast enhancement were observed between perfused and nonperfused segments. The wash-in effects of Gd-DTPA occurred during several seconds. Wilke et al,¹⁵⁶ using a turbo-FLASH sequence, studied the correlation between myocardial blood flow at rest and during dipyridamole-induced hyperemia in a closed chest dog model. They showed that contrast-enhanced ultrafast MR imaging allowed the assessment of myocardial perfusion both at rest and under stress circumstances. Wendland et al,¹⁵⁷ using echo-planar imaging in rats, injected the contrast agent gadodiamide and observed a 63% increase in signal in normal myocardial segments compared with ischemic segments. In a recent study, Saeed et al¹⁵⁸ showed in rats that the transit of the contrast agent gadodiamide, monitored by echo-planar imaging techniques, could be used to distinguish between reversibly and irreversibly injured myocardium. These experimental ultrafast MR imaging studies have stimulated the application of these techniques in patients with coronary artery disease.

Atkinson et al¹⁵⁵ and Van Rugge et al¹⁵⁹ evaluated the value of ultrafast MR imaging for the assessment of dynamic contrast enhancement and myocardial perfusion in healthy volunteers. Both studies showed progressively increasing signal intensities in the right ventricular cavity, the left ventricular cavity, and finally in the myocardial wall.

Manning et al¹⁶⁰ used ultrafast MR imaging for the assessment of myocardial perfusion abnormalities in patients with chest pain. Regional myocardium perfused by a diseased vessel demonstrated a lower peak signal intensity and lower rate of signal increase than did myocardium perfused by coronary arteries without stenosis. Repeat MR imaging study after revascularization showed an increase in peak signal intensity. The patients with an area of myocardium perfused by a diseased vessel and associated low peak signal intensity had the greatest improvement in regional peak signal intensity after revascularization. Van Rugge et al¹⁶¹ studied 20 patients with previous myocardial infarction using ultrafast Gd-DTPA MR imaging. After Gd-DTPA administration, infarcted myocardium demonstrated a signal intensity enhancement of 50%, whereas in normal myocardium an enhancement of 134% was obtained. Myocardial perfusion abnormalities were clearly observed in infarcted areas (Fig 8). The infarct site on MR imaging corresponded with the location of wall motion asynergy determined by echocardiography. Schaefer et al¹⁶² and Eichenberger et al¹⁶³ studied patients with coronary artery disease using Gd-DOTA–enhanced ultrafast MR imaging and dipyridamole stress. MR imaging showed a sensitivity, specificity, and accuracy of 65%, 76%, and 74%, respectively.¹⁶³ Based on these studies it can be concluded that contrast-enhanced ultrafast MR imaging allows noninvasive assessment of myocardial perfusion in patients with proven coronary artery disease.

View larger version (91K): [in this window] [in a new window]   Figure 8. Series of six sequential ultrafast magnetic resonance images after bolus administration of Gd-DTPA in a patient with healed myocardial infarction of the inferolateral wall. Ultimately (lower middle and lower right), decreased myocardial signal intensity is observed in the infarcted inferolateral wall compared with the normal anterior region (arrowheads). (From F.P. Van Rugge et al. Am J Cardiol. 1992;70:1233-1237. Reproduced with permission.)

To summarize, ultrafast MR imaging provides the opportunity to acquire dynamic information related to the passage of a paramagnetic contrast agent through the coronary circulation and thus provides an indirect measure of myocardial perfusion. Applying this technique to patients with the use of pharmacological stress with dipyridamole, myocardial regions perfused by a severely stenosed coronary artery can be detected by a delayed increase in signal intensity and a decreased peak signal intensity. A limitation is the acquisition of only one tomographic slice in most of these studies, whereas higher speed methods would allow multislice tomography to generate a three-dimensional perspective. In this way, ultrafast MR imaging can provide important information on the functional significance of coronary artery lesions.

Sequelae of Myocardial Infarction
MR imaging is capable of detecting short-term and long-term sequelae of acute myocardial infarction. Higgins et al¹⁶⁴ showed that segmental wall thinning was highly indicative of a sustained myocardial infarction in 9 of 10 patients with chronic infarctions. McNamara and Higgins¹⁶⁵ observed regional wall thinning in 20 of 22 patients with prior infarctions; in 10 of 14 patients with sufficient residual wall thickness for measurement of T2 relaxation times, decreased signal intensities and shortened T2 values were measured at the site of the infarcted area. In a study by Krauss et al,¹⁶⁶ 19 acute infarct patients were studied by MR imaging at discharge, of whom 13 patients were reexamined 4 to 7 months later. In 10 patients, infarct site and size did not change, and the T2 relaxation times remained prolonged, particularly in the patients with anterior infarction. The finding of prolonged T2 values in chronically infarcted areas was also observed in animal experiments by Checkley et al,¹⁶⁷ who found high-signal areas at 10 days in infarcted minipig hearts. After 2 weeks, no further change in signal intensity was detected, but myocardial thinning became more evident. Recently, Hsu et al¹⁶⁸ studied chronic myocardial infarcts (6 months or longer) in 10 formalin-fixed human autopsy hearts, and they showed significantly increased T1 and T2 values in infarcted tissue versus noninfarcted tissue. These studies indicate that detection of infarcted areas is possible at the chronic phase of infarction both by morphological appearance and altered signal intensities. Complications of acute myocardial infarction including ventricular aneurysm, ventricular septum perforation, mitral regurgitation, and left ventricular thrombus (Fig 9) also can be readily demonstrated by MR imaging.^{169 170} Left ventricular septal defect is clearly visualized by spin-echo MR imaging as absence of muscular tissue in the septal area, and by cine MR imaging as a signal void in the right ventricle.

View larger version (87K): [in this window] [in a new window]   Figure 9. Top, Transverse cardiac spin-echo magnetic resonance image plane showing a left ventricular thrombus occupying the whole left ventricle of a patient with a previously sustained large anterior wall infarction. Aneurysmal formation of the anteroapical area can be observed. After surgical removal, the largest diameter of the thrombus measured 7 cm. Bottom, Schematic drawing of top image. LV indicates left ventricle; RV, right ventricle; and T, thrombus.

To summarize, MR imaging provides an excellent means for detection of complications of myocardial infarction. Early detection of complications by MR imaging may be very important for guiding proper patient management.

Evaluation of Coronary Artery Bypass Grafts
MR imaging has been used to evaluate the patency of coronary artery bypass grafts. Using the spin-echo technique, the grafts appear as small circular structures with absence of luminal signal, since blood moves rapidly through normal grafts. However, sternal clips used in bypass grafting can lead to small regions of signal dropout that may be mistaken for patent grafts. There also must be sufficient flow to generate contrast between the graft lumen and the wall. Generally, multislice multiphase imaging is required to obtain the appropriate images for detecting rapid graft flow at contiguous levels in the same phase. In a study by Rubinstein et al¹⁷¹ using a multislice spin-echo technique in 20 patients after bypass surgery, the overall sensitivity and specificity for evaluating bypass patency were 92% and 85%, respectively. Gomes et al¹⁷² studied 20 patients with patent bypass grafts and showed that 54 of 64 grafts (84%) were detected by spin-echo MR imaging. Jenkins et al¹⁷³ assessed graft patency by spin-echo MR imaging in 22 patients and found 90% accuracy compared with contrast angiography. Frija et al¹⁷⁴ showed in 28 patients that spin-echo MR imaging after bypass surgery provided a correct diagnosis in 95% of cases. The major causes of diagnostic inaccuracies were hemostatic clips, in particular clips for internal mammary bypass grafts.

While the spin-echo technique shows lack of signal intensity in vascular compartments with rapid blood flow, the cine MR imaging technique depicts flowing blood as a bright signal. Therefore, the presence of a bright visible intraluminal signal is indicative of graft patency. First results of cine MR imaging by White et al¹⁷⁵ for determination of bypass patency in 25 patients showed accuracies of 91% for patency and 72% for occlusion. A subsequent study by White et al¹⁷⁶ in 10 patients showed for the determination of patency a sensitivity of 93%, a specificity of 86%, and an overall predictive accuracy of 89%. Aurigemma et al¹⁷⁷ used cine MR imaging in 20 operated patients with a total of 45 grafts and showed a sensitivity of 88%, a specificity of 100%, and an overall accuracy of 91%.

In summary, although these studies are preliminary, it has been presaged that a combined use of a spin-echo examination and cine MR imaging will be the optimal approach for imaging bypass grafts. Such a screening is applicable in postoperative chest pain syndromes to exclude graft occlusion or in screening for late graft occlusion or stenosis. Future flow-sensitive techniques are needed to exactly quantitate graft flow, as quantitation of bypass graft flow directly reflects distal runoff, which seems more valuable than simply detecting bypass patency.^{178 179 180 181}

Visualization of the Coronary Arteries
Noninvasive visualization of coronary arteries by MR imaging techniques may provide a tremendous tool for detecting stenoses in the main left and proximal coronary arteries in patients with coronary artery disease (Fig 10). However, these techniques are yet in an experimental phase and need further technical development. MR angiography is used routinely in many centers for evaluation of the carotid arteries and intracerebral vasculature, aortography, and assessment of the ileofemoral system. MR angiography of the coronary arteries, however, is technically more difficult because of the relatively small size of these arteries, their complex three-dimensional anatomy, and their constantly changing position within the thoracic cavity caused by cardiac motion and respiration. Several approaches for coronary MR angiography have been proposed. Initial attempts at MR imaging of the proximal coronary arteries had limited success because of the occurrence of artifact resulting from prominent cardiac and respiratory motion.^{182 183} More recently, other MR imaging techniques have been used to image the proximal coronary arteries in healthy subjects including MR imaging subtraction methods,^{184 185} three-dimensional MR angiograms formed by stacking two-dimensional planar images,¹⁸⁶ echo-planar imaging,¹⁸⁷ and fast spiral MR imaging.¹⁸⁸ Until now, the best results have been obtained with the use of an ultrafast MR angiographic technique during periods of breath holding.¹⁸⁹ Breath holding is essential to avoid excessive blurring from respiratory motion. MR imaging methods to reduce the bothersome signal related to fat ("fat suppression") are used to improve contrast between coronary arteries and surrounding epicardial fat. Transverse sections permit assessment of the left main, left anterior descending, and proximal right coronary arteries, whereas oblique imaging sections are best for depicting the left circumflex artery and the more distal segments of the right coronary artery. As many as 20 to 30 interleaved segments are acquired, and scan times (breath-holding periods) are 15 to 18 seconds for each image acquisition. In an initial study by Manning et al,¹⁹⁰ 19 normal subjects and 6 patients with coronary artery disease were imaged with the use of fat-suppressed breath-holding MR angiography (Fig 11). Imaging time was approximately 45 minutes. Mean vessel diameter visualized ranged from 2.6 mm (left circumflex artery) to 6.2 mm (left main coronary artery), which correlated with quantitative contrast angiography. Mean vessel length ranged from 8 mm (left main coronary artery) to 122 mm (right coronary artery). Occluded vessels appeared as absent flow signal distal to the occlusion, and high-grade stenoses appeared as signal loss in the area of stenosis with visualization of the vessel distally. In a subsequent study, Manning et al¹⁹¹ compared MR coronary angiography with conventional angiography. In this study in 39 subjects who were scheduled for elective cardiac catheterization with coronary angiography, the major epicardial coronary arteries were classified by MR angiography as being normal (or having minimal irregularities) or as having disease that was moderately severe to severe. The sensitivity and specificity of MR coronary angiography, as compared with conventional angiography, for correctly identifying individual vessels with 50% angiographic stenoses were 90% and 92%, respectively. The corresponding positive and negative predictive values were 85% and 95%, respectively. The sensitivity and specificity of the technique were 100% and 100%, respectively, for the left main coronary artery, 87% and 92% for the left anterior descending artery, 71% and 90% for the left circumflex coronary artery, and 100% and 78% for the right coronary artery. The entire procedure, during which both transverse and oblique imaging were performed, took about 20 minutes. Pennell et al¹⁹² studied 21 healthy control subjects and 5 patients with coronary artery disease with segmented k-space gradient-echo imaging, such that a complete image was obtained in 16 cardiac cycles during one breathhold. The left main stem (95%), left anterior descending coronary artery (91%), and right coronary artery were identified in all subjects, but identification of the left circumflex coronary artery was more difficult (76%). A good correlation was found between measurements made by MR imaging and contrast coronary angiography. Duerinckx and Urman¹⁹³ found a somewhat lower overall sensitivity for detecting significant coronary artery stenoses, which may be due to patient selection, independent evaluation, and differences in the technique used and experience in reading MR angiograms.

View larger version (69K): [in this window] [in a new window]   Figure 10. Top, Transverse magnetic resonance image of the origin of the right coronary artery (RCA) using a body coil. Bottom, Image quality is markedly improved with the surface coil. RAA indicates right atrial appendage.

View larger version (113K): [in this window] [in a new window]   Figure 11. Oblique magnetic resonance image along the major axis at the level of the proximal right coronary artery identified in transverse section. The image was acquired in breathhold using turbo fast low-angle shot imaging. (From W.J. Manning et al. Circulation. 1993;87:94-104.)

From these studies it was concluded that MR angiography provides a new approach to evaluate the patency of coronary arteries. It also was stated that MR imaging of the coronary arteries is at too early a stage to predict its future role. A present limitation of MR angiography is the requirement of a regular sinus rhythm and the need for breath holding during 15 to 18 seconds, although newer methods have already obviated the need for breath holding and possibly for a regular sinus rhythm.^{194 195 196} Besides, frequent ventricular extrasystoles result in the degradation of the quality of the image. The present spatial resolution and loss of signal due to turbulence preclude accurate prediction of stenosis severity. Recently it has been shown that it is not possible to identify stenotic vessels based on quantification of signal intensity¹⁹⁷ ; no significant difference in signal was found between vessel segments of a normal coronary artery and vessel segments proximal to a significant stenosis. The use of faster and stronger gradient coils and improved surface coils may improve spatial resolution and the signal-to-noise ratio in the future. Poncelet et al¹⁹⁸ demonstrated the potential of echo-planar MR imaging to detect flow velocity changes in coronary arteries during isometric exercise. These findings suggest the utility of this technique to evaluate coronary flow reserve. It is assumed that if all these advances can be made, MR imaging may become useful for screening the major coronary arteries for significant coronary artery disease.

	Conclusions

Top Abstract Introduction MR Imaging Techniques Coronary Artery Disease Conclusions References

 
At present, MR imaging provides useful information that is not readily available from other noninvasive conventional modalities such as echocardiography, radionuclide angiography, and computed tomography (Table 2). The superb resolution, the inherent contrast, the three-dimensional nature, the lack of ionizing radiation, and its morphological imaging capabilities sufficiently justify the application of MR imaging in clinical cardiology.

View this table: [in this window] [in a new window]   Table 2. Comparison of Clinical Utility of Noninvasive Imaging Methods

At present, MR techniques allow the evaluation of anatomy and function (accepted use), perfusion and viability (development phase), and coronary angiography (experimental phase). A particular strength of MR imaging is that one single cardiac MR test may encompass cardiac anatomy, perfusion, function, metabolism, and coronary angiography. Consequently, MR imaging has greater potential than any diagnostic instrument yet conceived. The expected replacement of multiple diagnostic tests such as echocardiography, nuclear medicine procedures, and diagnostic contrast arteriography with one MR test may have profound effects on cardiovascular healthcare economics. The definite judgments about the relative importance of MR imaging as a valuable clinical diagnostic tool must be settled. For those situations in which MR imaging techniques can replace the conventional techniques, these judgments should be based on large prospective multicenter studies. For MR imaging to have its most substantial impact in detecting coronary artery disease, future technical developments should allow definition of accurate distribution of regional myocardial blood flow during stress in order to assess the ischemic area at risk and the visualization of the coronary arteries with quantitation of coronary flow at multiple cardiac levels. These advances include faster imaging sequences, automated quantification algorithms, three-dimensional angiography, and the development of conventional exercise devices. Many of these improvements are already developed or still under development. If these technical advances will be applied on a large scale, then MR imaging may be primarily useful for screening the major coronary arteries for significant coronary artery disease. In addition, MR coronary arteriography could be used clinically in screening for abnormal origin of the coronary arteries, in diagnosing coronary artery disease in patients who present with chest pain or other suggestive symptoms, in monitoring the progression of disease in patients with known coronary artery disease, and in making decisions about treatment. Because of its safety, it could be used in younger age groups and patients with contraindications to conventional contrast angiography. Particularly, early detection and flow assessment of stenosed coronary arteries by MR angiography using flow velocity will constitute a tremendous progress in clinical cardiology that would far outweigh the cost inherent to the MR imaging procedure.

	Acknowledgments

 
Our secretary, Mrs A. van der Mey, is gratefully acknowledged for carefully typing the manuscript.

	References

Top Abstract Introduction MR Imaging Techniques Coronary Artery Disease Conclusions References

 

1. Kaufman L, Crooks L, Sheldon P, Hricak H, Herfkens R, Bank W. The potential impact of nuclear magnetic resonance imaging on cardiovascular diagnosis. Circulation. 1983;67:251-257. [Abstract/Free Full Text]
   
2. Herfkens RJ, Higgins CB, Hricak H, Lipton MJ, Crooks LE, Lanzer P, Botvinick E, Brundage B, Sheldon PE, Kaufman L. Nuclear magnetic resonance imaging of the cardiovascular system: normal and pathologic findings. Radiology. 1983;147:749-759. [Abstract/Free Full Text]
   
3. Van der Wall EE, De Roos A, Van Voorthuisen AE, Bruschke AVG. Magnetic resonance imaging: a new approach for evaluating coronary artery disease? Am Heart J. 1991;121:1202-1220.
   
4. De Roos A, Van der Wall EE. Magnetic resonance imaging and spectroscopy of the heart. Curr Opin Cardiol. 1991;6:946-952. [Medline] [Order article via Infotrieve]
   
5. Von Schulthess GK, Fisher MR, Crooks LE, Higgins CB. Gated MR imaging of the heart: intracardiac signal in patients and healthy subjects. Radiology. 1985;157:125-132.
   
6. Mohiaddin RH, Longmore DB. Functional aspects of cardiovascular nuclear imaging: techniques and application. Circulation. 1993;88:264-281.[Free Full Text]
   
7. Pattynama PMT, De Roos A, Van der Wall EE, Van Voorthuisen AE. Evaluation of cardiac function with magnetic resonance imaging. Am Heart J. 1994;128:595-607. [Medline] [Order article via Infotrieve]
   
8. Cranney GB, Pohost GM. Nuclear magnetic resonance of the cardiovascular system. Curr Opin Cardiol. 1988;3:958-969.
   
9. Bottomley PA. Noninvasive study of high-energy phosphate metabolism in the human heart by depth-resolved 31P NMR spectroscopy. Magn Reson Imaging. 1992;2:711-719.
   
10. Florentine MS, Grosskreutz CL, Chang W, Hartnett JA, Dunn VD, Ehrhardt JC, Fleagle SR, Collins SM, Marcus ML, Skorton DJ. Measurement of left ventricular mass in vivo using gated nuclear magnetic resonance imaging. J Am Coll Cardiol. 1986;8:107-112. [Abstract]
    
11. Keller AM, Peshock RM, Malloy CR, Buja LM, Nunnaly R, Parkey RW, Willerson JT. In vivo measurement of myocardial mass using nuclear magnetic resonance imaging. J Am Coll Cardiol. 1986;8:113-117. [Abstract]
    
12. Caputo GR, Tscholakoff D, Sechtem U, Higgins CB. Measurements of canine left ventricular mass by using MR imaging. AJR Am J Roentgenol. 1987;148:33-38. [Abstract/Free Full Text]
    
13. Markiewicz W, Sechtem U, Kirby R, Derugin N, Caputo GC, Higgins CB. Measurement of ventricular volumes in the dog by nuclear magnetic resonance imaging. J Am Coll Cardiol. 1987;10:170-177. [Abstract]
    
14. Katz J, Milliken MC, Stray-Gundersen J, Buja LM, Parkey RW, Mitchell JH, Peshock RM. Estimation of human myocardial mass with MR imaging. Radiology. 1988;169:495-498. [Abstract/Free Full Text]
    
15. Ostrzega E, Maddahi J, Honma H, Crues JV III, Resser KJ, Charuzi Y, Berman DS. Quantification of left ventricular myocardial mass in humans by nuclear magnetic resonance imaging. Am Heart J. 1989;117:444-452. [Medline] [Order article via Infotrieve]
    
16. Friedman BJ, Waters J, Kwan OL, DeMaria AN. Comparison of magnetic resonance imaging and echocardiography in determination of cardiac dimensions in normal subjects. J Am Coll Cardiol. 1985;5:1369-1376. [Abstract]
    
17. Longmore DB, Klipstein RH, Underwood SR, Firman DN, Hounsfield GN, Watanabe M, Bland C, Fox K, Poole-Wilson PA, Rees RSO, Denison D, McNeilly AM, Burman ED. Dimensional accuracy of magnetic resonance in studies of the heart. Lancet. 1985;1:1360-1362. [Medline] [Order article via Infotrieve]
    
18. Byrd BF III, Schiller NB, Botvinick EH, Higgins CB. Normal cardiac dimensions by magnetic resonance imaging. Am J Cardiol. 1985;55:1440-1442. [Medline] [Order article via Infotrieve]
    
19. Shapiro EP, Rogers WJ, Beyar R, Soulen RL, Zerhouni EA, Lima JAC, Weiss JL. Determination of left ventricular mass by magnetic resonance imaging in hearts deformed by acute infarction. Circulation. 1989;79:706-711. [Abstract/Free Full Text]
    
20. Møgelvang J, Thomsen C, Mehlsen J, Bräckle G, Stubgaard M, Henriksen O. Evaluation of left ventricular volumes measured by magnetic resonance imaging. Eur Heart J. 1986;7:1016-1021. [Abstract/Free Full Text]
    
21. Van Rossum AC, Visser FC, Van Eenige MJ, Valk J, Roos JP. Magnetic resonance imaging of the heart for determination of ejection fraction. Int J Cardiol. 1988;18:53-63. [Medline] [Order article via Infotrieve]
    
22. Van Rossum AC, Visser FC, Sprenger M, Van Eenige MJ, Valk J, Roos JP. Evaluation of magnetic resonance imaging for determination of left ventricular ejection fraction and comparison with angiography. Am J Cardiol. 1988;62:628-633. [Medline] [Order article via Infotrieve]
    
23. Just H, Holubarsch C, Friedburg H. Estimation of left ventricular volume and mass by magnetic resonance imaging: comparison with quantitative biplane angiocardiography. Cardiovasc Intervent Radiol. 1987;10:1-4. [Medline] [Order article via Infotrieve]
    
24. Underwood SR, Rees RSO, Savage PE, Klipstein RH, Firmin DN, Fox KM, Poole-Wilson PA, Longmore DB. Assessment of regional left ventricular function by magnetic resonance. Br Heart J. 1986;56:334-340. [Abstract/Free Full Text]
    
25. Fisher MR, von Schulthess GK, Higgins CB. Multiphasic cardiac magnetic resonance imaging: normal regional left ventricular wall thickening. AJR Am J Roentgenol. 1985;145:27-30. [Abstract/Free Full Text]
    
26. Sechtem U, Sommerhoff BA, Markiewicz W, White RD, Cheitlin MD, Higgins CB. Regional left ventricular wall thickening by magnetic resonance imaging: evaluation in normal persons and patients with global and regional dysfunction. Am J Cardiol. 1987;59:145-151. [Medline] [Order article via Infotrieve]
    
27. Akins EW, Hill JA, Sieve