Contrast-Enhanced Cardiac Magnetic Resonance Imaging




Cardiac magnetic resonance (CMR) imaging has significantly evolved in the past decade and is well established in the evaluation of coronary artery disease (CAD). The evaluation of cardiac anatomy and contractility by high-resolution CMR can be improved by using intravenous administration of gadolinium-based contrast agents. Delayed enhancement CMR imaging has become the gold standard for quantification of myocardial viability in CAD. Contrast-enhanced CMR imaging may circumvent the need for endomyocardial biopsy or localize the involved regions, thereby improving the diagnostic yield of this invasive procedure. The application of contrast-enhanced CMR as an advanced imaging technique for ischemic and nonischemic diseases is reviewed.








  • Clinical assessment of myocardial perfusion remains critical in determining the diagnosis, management, and prognosis of patients with suspected or known coronary artery disease (CAD).



  • Multiparametric cardiac magnetic resonance (CMR) (stress perfusion, rest perfusion, and late gadolinium enhancement) has better sensitivity and negative predictive value (NPV) than single-photon emission CT (SPECT) for the diagnosis of CAD.



  • Current guidelines and appropriate use criteria indicate that coronary artery magnetic resonance (MR) angiography is useful for identifying coronary artery anomalies, in particular in younger individuals, without exposure to ionizing radiation or iodinated contrast medium.



  • Delayed enhancement CMR, one of the most common examinations for tissue characterization both in ischemic and nonischemic myocardial diseases, has become the gold standard for visualization and quantification of infarcted myocardium and scar tissues as well as for the detection of infiltrative diseases of the heart.



Key Points


Introduction


Despite a significant decline in the death rate attributable to cardiovascular diseases in recent years, they are still responsible for 1 in every 3 deaths (32.8%) in the United States. More than 2200 Americans die each day from cardiovascular diseases (1 every 39 seconds). CAD causes 1 of every 6 deaths, with more than 400,000 deaths annually. An estimate of 785,000 Americans have a new coronary artery attack and 470,000 have a recurrent coronary artery attack each year. Additionally, approximately 195,000 silent first myocardial infarctions (MIs) occur annually. All this together means a coronary event every 25 seconds with a death rate of 1 every minute.




Myocardial perfusion imaging


Clinical assessment of myocardial perfusion remains critical in determining the diagnosis, management, and prognosis of patients with suspected or known CAD. Even though catheter angiography and cardiac-gated CT angiography are excellent modalities to demonstrate the patency of coronary arteries, they tell little about the downstream microvascular flow within the myocardium. Myocardial ischemia is detected in fewer than half of patients with obstructive CAD; and 10% of patients with normal angiogram and low to intermediate probability of CAD have abnormal myocardial perfusion. Because perfusion abnormalities proceed systolic dysfunction, there is no surprise that direct perfusion imaging has higher sensitivity than indirect imaging (eg, wall motion dysfunction) for detection of ischemia, a concept described as “ischemic cascade” ( Fig. 1 ).




Fig. 1


The ischemic cascade refers to the temporal sequence of events that develop over time when there is a progressive imbalance between myocardial oxygen demand and supply with abnormal myocardial perfusion being the first detectable event.


Stress Echocardiography and Perfusion Scintigraphy


Stress echocardiography can evaluate myocardial perfusion by detecting wall motion abnormalities in response to physical (exercise) or pharmacologic (mainly dobutamine or dipyridamole) stress. The sensitivity and specificity of exercise echocardiography range from 74% to 97% and 64% to 86%, respectively. For dobutamine stress echocardiogram, the sensitivity and specificity are 61% to 95% and 51% to 95%. Dipyridamole stress echocardiogram has sensitivity and specificity in the range of 61% to 81% and 90% to 94%, respectively. Like perfusion scintigraphy, the diagnostic performance of stress echocardiography is decreased in multivessel disease. Although stress echocardiography is a versatile tool that does not involve radiation exposure, interpretation is mainly qualitative and based on the visual assessment of wall motion (thickening). Diagnostic performance is operator dependent with moderate interobserver agreement of 73% and κ value of 0.37. Additionally, finding of an appropriate acoustic window can be challenging in some cases. Sympathomimetic effect of dobutamine may induce hypotension, headache, anxiety, and arrhythmias. Contraindications of dobutamine administration include several conditions that are common in patients with cardiovascular disease (eg, severe arterial hypertension, unstable angina, significant aortic stenosis, complex cardiac arrhythmias, hypertrophic obstructive cardiomyopathy, myocarditis, uncontrolled heart failure, and history of hypersensitivity to the medication). Dobutamine stress CMR has also been extensively used for detection of inducible ischemia with good sensitivity (79%–96%) and specificity (70%–90%); but many of the limitations (discussed previously) for stress echocardiography (eg, subjective visual assessment and contraindication to the use of dobutamine) also apply to this technique. This imaging modality is also limited in patients with moderate to severe reduction in ejection fraction and in those with left ventricular hypertrophy. More recently, real-time contrast echocardiography using encapsulated microbubbles has been introduced for evaluation of myocardial perfusion. Limitations include low spatial resolution with inadequate endocardial border definition, limited acoustic windows, and concerns related to potential mechanical obstruction of the coronary vasculature by microbubbles. Contraindications to the use of microbubbles include intra-arterial injection, intracardiac shunt, unstable heart failure, acute MI or coronary syndrome, ventricular arrhythmias, respiratory failure, pulmonary hypertension and hypersensitivity to perflutren, blood products, or albumin.


Nuclear cardiac imaging with SPECT using 201 TI-labeled or 99m Tc-labeled agents is probably the most widely used noninvasive imaging technique for evaluation of myocardial perfusion. Cardiac nuclear imaging, however, exposes patients to a significant dose of ionizing radiation and has important limitations, including poor spatial resolution, inability to perform quantitative measurements, and susceptibility to attenuation artifacts. In addition, the acquisition time for stress-induced myocardial ischemia is lengthy, and it usually requires the stress and rest portions of the study to be performed in separate sessions. More recently, positron emission tomography (PET) with different tracers (ammonia N 13, water O 15, or rubidium chloride Rb 82) has proved useful for the quantitative measurement of myocardial blood flow and coronary flow reserve. PET also has significant shortcomings, however: it is not widely available, it is expensive, and, because of the short half-life of the radiotracers used for perfusion imaging (ammonia N 13, 9.8 minutes; water O 15, 2.4 minutes; and rubidium chloride Rb 82, 78 seconds), it requires a cyclotron on site.


The NPV of exercise myocardial perfusion scintigraphy or echocardiography is high. Several different meta-analysis, including many studies with thousands of patients, have demonstrated the value, cost-effectiveness, and safety of myocardial perfusion scintigraphy and stress echocardiography for the diagnosis of CAD. A meta-analysis that included 17 nuclear medicine perfusion studies (8008 patients) and 4 exercise echocardiography studies (3021 patients) reported greater than 98% NPV for MI and cardiac death over a 36-month follow up. In a different meta-analysis, the pooled annualized event rate for cardiac death and MI are substantially (5-fold) higher in patients with abnormal stress perfusion imaging. A more recent meta-analysis in hypertensive patients found sensitivity of 90% and 77% and specificity of 63% and 89% for myocardial perfusion scintigraphy and stress echocardiography, respectively, in this population at risk for CAD.


Magnetic Resonance Myocardial Perfusion Imaging


CMR imaging is a multiparametric study, given its ability to assess multiple aspects of cardiovascular pathology in a single examination, including myocardial and coronary artery anatomy, ventricular function, myocardial perfusion, and viability. First-pass contrast-enhanced MR imaging has emerged as an excellent alternative imaging modality for the assessment of myocardial perfusion. The higher contrast and spatial/temporal resolution of CMR imaging compared with other techniques not only allow a more accurate detection of ischemia and CAD but also provide detailed anatomic and functional information.


CMR has evolved significantly in the past decade and has finally established its place in the menu of imaging technologies available in the evaluation of CAD. In the European Cardiovascular Magnetic Resonance Registry pilot study, conducted in 20 German centers, in which a total of 11,400 patients were included, 88% of patients received a gadolinium-based contrast and 21% underwent adenosine stress perfusion. Image quality was considered good (90%) or moderate (8%) in the vast majority of studies performed. Severe complications were rare (0.05%), with no death during CMR examinations. In two-thirds of patients, the findings of CMR imaging had an impact on clinical management; the final diagnosis was entirely different in 16% of cases, which resulted in a complete change in management. And finally, in more than 86% of cases, CMR alone satisfied all imaging needs for patient care, so that no additional imaging was required. The American College of Cardiology, American College of Radiology, American Heart Association, North American Society for Cardiovascular Imaging, and Society for Cardiovascular Magnetic Resonance Imaging Expert Consensus Document on Cardiovascular Magnetic Resonance indicates that the use of CMR stress testing (vasodilator or dobutamine) is appropriate in individuals with intermediate pretest probability of CAD, patients with an uninterpretable EEG, or those unable to exercise. CMR stress testing is also appropriate for cardiac risk assessment in patients with prior coronary angiography or stenosis of unclear clinical significance.


Clinical Validation of MR Perfusion Imaging


Many single-center studies have demonstrated excellent performance of CMR for the detection of CAD compared with other imaging modalities. The few multicenter studies that have been performed, which probably better reflect the real world than specialized single-center studies, have similarly demonstrated the superiority of CMR over other imaging techniques. In a large prospective trial, including 752 patients, in which CMR and SPECT were compared with x-ray coronary angiography as the reference standard, multiparametric CMR (stress perfusion, rest perfusion, and late gadolinium enhancement) had better sensitivity than SPECT (86.5% vs 66.5%) and NPV (90.5% vs 79.1%) for diagnosis of CAD. When considered alone, the stress perfusion part of the examination also out-performed SPECT both in the single-vessel and the multivessel disease groups. Stress myocardial perfusion MR also helps with risk stratification of patients with known coronary artery stenosis of intermediate angiographic severity and may help identification of patients at risk of major adverse cardiovascular event (ie, death, stroke, and MI). Patients with coronary lesions of intermediate severity (50%–75% diameter stenosis) with myocardial perfusion defect have significantly higher incidence of adverse outcome (4%–20%) than those with equivalent coronary artery stenosis without perfusion abnormality. In addition, a normal stress test is associated with low event rate. In a series of 513 patients with known or suspected CAD, those with normal CMR stress test (adenosine stress perfusion and dobutamine stress wall motion) had a 3-year event-free survival rate greater than 99%.


Two different recent meta-analyses have reported CMR perfusion imaging to have sensitivity of 89% to 91% and specificity of 80% to 81% for the detection of CAD. The exact role of myocardial perfusion MR imaging in patients with acute chest pain and acute coronary syndrome presenting to the emergency department is still to be defined. Some studies have demonstrated, however, that CMR perfusion imaging is feasible in these patients and has high diagnostic accuracy for detection of true acute coronary syndromes, particularly when adenosine vasodilator stress is used (sensitivity: 77%–100%, specificity: 83%–93%, and accuracy: 87%). Follow-up after normal perfusion CMR examinations has demonstrated an excellent NPV of 94% to 100% for adverse outcome or subsequent diagnosis of CAD after hospital discharge. An imaging strategy involving perfusion CMR may also reduce cost in patient care by reducing unnecessary admissions and cardiac catheterizations without missing true positive acute coronary syndromes. It remains unclear how versatile and cost-effective CMR would be compared with other imaging modalities (ie, SPECT) used for early detection of acute coronary syndrome in the emergency department.


Imaging Technique


First-pass perfusion CMR examination is based on dynamic rapid imaging of the heart during the circulation of a gadolinium-based contrast agent from the cardiac cambers into the myocardium. CMR perfusion imaging can be performed as a stand-alone technique or more commonly as part of a comprehensive CMR protocol (multiparametric examination), which includes noncontrast acquisition for morphologic and functional assessment as well as delayed enhancement (DE) images. Different protocols exist for comprehensive examination of the heart, but all contain morphologic and functional sequences; a stress dynamic first-pass perfusion after contrast administration, which is compared with a similar acquisition at rest; and finally delayed images for the detection of late gadolinium enhancement. Many variations exist depending on the type of magnet, coils, clinical question at hand, stressor agent used, and personal preferences.


A typical MR protocol begins with localizers that are used to determine the true left ventricular short and long axes. These localizers are usually obtained as single-shot technique either as half-Fourier acquisition single-shot turbo spin-echo (HASTE) or as steady-state free precession (SSFP). Subsequently, functional images with white blood techniques are obtained in the short axis from the mitral valve through the apex, in vertical long axis, and in horizontal long axis views. Next, perfusion MR is obtained during the first pass of a gadolinium-based contrast agent after intravenous injection during pharmacologic vasodilation with adenosine or dipyridamole. Approximately 10 to 15 minutes later, allowing for contrast media elimination from the circulation, rest imaging is performed in the same plane and with an identical sequence (short-axis saturation recovery). Finally myocardial viability and infarction are evaluated with DE technique in which a heavily T1-weighted segmented gradient-recalled echo (GRE) sequence is acquired in at least 2 planes, usually short axis and vertical long axis. Some investigators prefer to perform the stress myocardial perfusion study first, followed by MR coronary angiography and resting myocardial perfusion study. Subsequently, additional intravenous injection of gadolinium is administered in preparation for delayed images. In the meantime, resting wall images are obtained, with DE images acquired at the end. Other investigators prefer to perform the resting perfusion examination first, followed by cine images, MR angiography, and subsequently the stress perfusion and viability examination performed at the end of the exam. A stress-only protocol has been proposed based on the high diagnostic performance of the hyperemia data from different studies. Adding T2-weighted imaging for depiction of edema has also been proposed for better detection of patients with acute coronary syndrome. The order in which stress and rest perfusion are performed may also be influenced by whether adenosine or dipyridamole is used as the stressor agent. Because adenosine has a short half-life (<10 seconds), it may be better to perform the stress study first so that the rest perfusion examination is not influenced by residual gadolinium injected at rest, which may accumulate in areas of scar tissue. If dipyridamole is used, it may be better to perform the rest study first, and then the stress study, because it has a significant longer half-life (30 minutes) and may delay the completion of the study ( Fig. 2 ).




Fig. 2


( A ) Typical first-pass perfusion protocol for adenosine stress perfusion CMR. ( B ) Typical first-pass perfusion protocol for dipyridamole stress perfusion CMR. ( C ) Typical first-pass perfusion protocol for regadenoson stress perfusion CMR.

( Modified from Gerber BL, Raman SV, Nayak K, et al. Myocardial first-pass perfusion cardiovascular magnetic resonance: history, theory, and current state of the art. J Cardiovasc Magn Reson 2008;10:18.)


Stress Agents


Adenosine is a naturally occurring substance that activates A 2 cell-surface receptors of vascular smooth muscle cells causing relaxation and hence vasodilation. Unlike normal vessels or mildly abnormal arteries, the more diseased coronary arteries have a reduced blood flow reserve and cannot further dilate in response to adenosine, which creates a heterogeneous flow pattern across the myocardium. The vasodilatory effect may create mild to moderate reduction in systolic and diastolic blood pressure and reflex tachycardia. Significant advantages of adenosine over other stress agents used in cardiovascular imaging are its short half-life and excellent safety profile, with rare serious complications. Contraindications of adenosine include second-degree and third-degree atrioventricular block, sick sinus syndrome, symptomatic bradycardia, severe asthma, and chronic obstructive pulmonary disease.


Dipyridamole is an indirect coronary vasodilator that inhibits intracellular reuptake and deamination of adenosine, increasing intravascular adenosine levels. Regadenoson, a selective A 2A adenosine receptor agonist, is a newly introduced stressor agent that induces functional and perfusion results similar to nonselective adenosine and is used for stress perfusion imaging examinations. Regadenoson has several advantages over adenosine: it is administered as an intravenous bolus fixed dose (400 μg independent of the patient weight); has fewer adverse effects, including atrioventricular block and bronchospasm; can be used in patients with mild to moderate reactive airway disease; and is cheaper. Patients should be advised to restrain from smoking, drinking tea, coffee, or any caffeine-containing substance at least for 12 hours (some prefer up to 48 hours) previous to the examination due to the competitive interaction between caffeine and adenosine or regadenoson, which may attenuate the coronary hyperemic response by competitive blockade of A 2A receptors. For stress perfusion imaging, intravenous adenosine is administered at a dose of 140 μg/kg/min. Approximately 4 minutes into the adenosine infusion, an intravenous bolus injection of 0.05 mmol/kg to 0.01 mmol/kg of gadolinium-based contrast at 3 mL/min to 4 mL/min is administered, followed by 15 mL of saline flush at 5 mL/s, during end-expiratory breath hold.


The majority of clinical trials published so far have been done with 1.5-T magnets but the 3-T magnets are being used more and more in the clinical arena. Imaging at 3 T differs significantly from 1.5 T because it suffers from increased susceptibility artifacts, difference in tissue relaxation, and radiofrequency homogeneity problems. At the same time, higher field strength increases the signal-to-noise ratio, which theoretically improves image quality and may reduce imaging time. Only a few studies of myocardial perfusion performed by 3-T magnets have been published, but so far they tend to demonstrate a better contrast and spatial resolution, with improved accuracy in diagnosis of myocardial perfusion defects, with sensitivity as high as of 90%.


Image Interpretation


Qualitative interpretation of perfusion CMR imaging is the most common approach in clinical practice. Image interpretation usually begins with review of delayed images for the presence of scar from previous MI. In absence of scar, the nonischemic myocardium exhibits uniform enhancement during first-pass perfusion at rest and with vasodilator stress. Ischemic but noninfarcted myocardium shows perfusion abnormality at stress that normalizes at rest ( Fig. 3 ). Matched stress and rest perfusion defects in the absence of DE in the same region are considered artifactual. Quantitative analysis of CMR perfusion examination is possible but complex. It can be significantly affected by several artifacts and variables (timing of saturation pulse, concentration of the contrast agent, magnetic field strength, phase of the cardiac cycle, and so forth) and has not been entirely validated or standardized. Therefore, quantitative analysis is more commonly used in research than in clinical practice.




Fig. 3


Stress perfusion CMR. ( A ) Stress image shows subendocardial hypoperfusion of the interventricular septum ( arrow ). ( B ) Corresponding rest image reveals normal enhancement of the ischemic zone ( arrow ).


A common artifact that is seen in as many as 1 in every 4 studies is the endocardial dark rim artifact, which can be seen in dynamic sequences when the bolus of gadolinium first enters into the left ventricle and mimics a perfusion defect ( Fig. 4 ). This dark rim is considered an artifact from susceptibility effect from gadolinium itself, from motion, or from resolution effect, because the artifact is perpendicular to the direction with the lowest spatial resolution (typically the phase-encoding direction). The most common location is at the interventricular septum in the left anterior descending artery vascular territory between the right and left ventricular cavities (>50%). Unlike true perfusion defects, this artifact is transient and does not remain after the bolus of contrast, a feature that helps in their differentiation.




Fig. 4


Dark rim artifact versus perfusion defect. ( A ) The dark rim artifact is seen as a thin endocardial low signal region ( white arrows ). ( B ) Perfusion defect in the interventricular septum ( black arrow ). The dark rim artifact is usually transient whereas real perfusion defect persist longer during the first pass of the bolus of contrast.


Limitations and Contraindications


Contraindications of perfusion CMR can be related to the high magnetic field, contrast agent, or pharmacologic stressor used. Contraindications related to the magnetic field are the same as those for other MR examination, including patients with electronic devices, such as most pacemakers and defibrillators, neurostimulators, ear implants, and cerebral aneurysm clips. These are usually considered absolute contraindications. Currently, the use of gadolinium contrast agents for cardiac imaging is off-label in the United States. Severe anaphylactic reaction to gadolinium is rare. A contraindication to the use of intravascular gadolinium is renal insufficiency because of the risk of nephrogenic systemic fibrosis, especially with the use of linear gadolinium chelates. Finally, contraindication for the use of adenosine or dipyridamole include asthma, severe chronic obstructive pulmonary disease, unstable angina, severely high or low blood pressure, heart block or sick sinus syndrome, and severe heart failure. Claustrophobia, seen in approximately 5% of patients, is a relative contraindication that can be modified with the use of superficial sedatives or tranquilizers.




Myocardial perfusion imaging


Clinical assessment of myocardial perfusion remains critical in determining the diagnosis, management, and prognosis of patients with suspected or known CAD. Even though catheter angiography and cardiac-gated CT angiography are excellent modalities to demonstrate the patency of coronary arteries, they tell little about the downstream microvascular flow within the myocardium. Myocardial ischemia is detected in fewer than half of patients with obstructive CAD; and 10% of patients with normal angiogram and low to intermediate probability of CAD have abnormal myocardial perfusion. Because perfusion abnormalities proceed systolic dysfunction, there is no surprise that direct perfusion imaging has higher sensitivity than indirect imaging (eg, wall motion dysfunction) for detection of ischemia, a concept described as “ischemic cascade” ( Fig. 1 ).




Fig. 1


The ischemic cascade refers to the temporal sequence of events that develop over time when there is a progressive imbalance between myocardial oxygen demand and supply with abnormal myocardial perfusion being the first detectable event.


Stress Echocardiography and Perfusion Scintigraphy


Stress echocardiography can evaluate myocardial perfusion by detecting wall motion abnormalities in response to physical (exercise) or pharmacologic (mainly dobutamine or dipyridamole) stress. The sensitivity and specificity of exercise echocardiography range from 74% to 97% and 64% to 86%, respectively. For dobutamine stress echocardiogram, the sensitivity and specificity are 61% to 95% and 51% to 95%. Dipyridamole stress echocardiogram has sensitivity and specificity in the range of 61% to 81% and 90% to 94%, respectively. Like perfusion scintigraphy, the diagnostic performance of stress echocardiography is decreased in multivessel disease. Although stress echocardiography is a versatile tool that does not involve radiation exposure, interpretation is mainly qualitative and based on the visual assessment of wall motion (thickening). Diagnostic performance is operator dependent with moderate interobserver agreement of 73% and κ value of 0.37. Additionally, finding of an appropriate acoustic window can be challenging in some cases. Sympathomimetic effect of dobutamine may induce hypotension, headache, anxiety, and arrhythmias. Contraindications of dobutamine administration include several conditions that are common in patients with cardiovascular disease (eg, severe arterial hypertension, unstable angina, significant aortic stenosis, complex cardiac arrhythmias, hypertrophic obstructive cardiomyopathy, myocarditis, uncontrolled heart failure, and history of hypersensitivity to the medication). Dobutamine stress CMR has also been extensively used for detection of inducible ischemia with good sensitivity (79%–96%) and specificity (70%–90%); but many of the limitations (discussed previously) for stress echocardiography (eg, subjective visual assessment and contraindication to the use of dobutamine) also apply to this technique. This imaging modality is also limited in patients with moderate to severe reduction in ejection fraction and in those with left ventricular hypertrophy. More recently, real-time contrast echocardiography using encapsulated microbubbles has been introduced for evaluation of myocardial perfusion. Limitations include low spatial resolution with inadequate endocardial border definition, limited acoustic windows, and concerns related to potential mechanical obstruction of the coronary vasculature by microbubbles. Contraindications to the use of microbubbles include intra-arterial injection, intracardiac shunt, unstable heart failure, acute MI or coronary syndrome, ventricular arrhythmias, respiratory failure, pulmonary hypertension and hypersensitivity to perflutren, blood products, or albumin.


Nuclear cardiac imaging with SPECT using 201 TI-labeled or 99m Tc-labeled agents is probably the most widely used noninvasive imaging technique for evaluation of myocardial perfusion. Cardiac nuclear imaging, however, exposes patients to a significant dose of ionizing radiation and has important limitations, including poor spatial resolution, inability to perform quantitative measurements, and susceptibility to attenuation artifacts. In addition, the acquisition time for stress-induced myocardial ischemia is lengthy, and it usually requires the stress and rest portions of the study to be performed in separate sessions. More recently, positron emission tomography (PET) with different tracers (ammonia N 13, water O 15, or rubidium chloride Rb 82) has proved useful for the quantitative measurement of myocardial blood flow and coronary flow reserve. PET also has significant shortcomings, however: it is not widely available, it is expensive, and, because of the short half-life of the radiotracers used for perfusion imaging (ammonia N 13, 9.8 minutes; water O 15, 2.4 minutes; and rubidium chloride Rb 82, 78 seconds), it requires a cyclotron on site.


The NPV of exercise myocardial perfusion scintigraphy or echocardiography is high. Several different meta-analysis, including many studies with thousands of patients, have demonstrated the value, cost-effectiveness, and safety of myocardial perfusion scintigraphy and stress echocardiography for the diagnosis of CAD. A meta-analysis that included 17 nuclear medicine perfusion studies (8008 patients) and 4 exercise echocardiography studies (3021 patients) reported greater than 98% NPV for MI and cardiac death over a 36-month follow up. In a different meta-analysis, the pooled annualized event rate for cardiac death and MI are substantially (5-fold) higher in patients with abnormal stress perfusion imaging. A more recent meta-analysis in hypertensive patients found sensitivity of 90% and 77% and specificity of 63% and 89% for myocardial perfusion scintigraphy and stress echocardiography, respectively, in this population at risk for CAD.


Magnetic Resonance Myocardial Perfusion Imaging


CMR imaging is a multiparametric study, given its ability to assess multiple aspects of cardiovascular pathology in a single examination, including myocardial and coronary artery anatomy, ventricular function, myocardial perfusion, and viability. First-pass contrast-enhanced MR imaging has emerged as an excellent alternative imaging modality for the assessment of myocardial perfusion. The higher contrast and spatial/temporal resolution of CMR imaging compared with other techniques not only allow a more accurate detection of ischemia and CAD but also provide detailed anatomic and functional information.


CMR has evolved significantly in the past decade and has finally established its place in the menu of imaging technologies available in the evaluation of CAD. In the European Cardiovascular Magnetic Resonance Registry pilot study, conducted in 20 German centers, in which a total of 11,400 patients were included, 88% of patients received a gadolinium-based contrast and 21% underwent adenosine stress perfusion. Image quality was considered good (90%) or moderate (8%) in the vast majority of studies performed. Severe complications were rare (0.05%), with no death during CMR examinations. In two-thirds of patients, the findings of CMR imaging had an impact on clinical management; the final diagnosis was entirely different in 16% of cases, which resulted in a complete change in management. And finally, in more than 86% of cases, CMR alone satisfied all imaging needs for patient care, so that no additional imaging was required. The American College of Cardiology, American College of Radiology, American Heart Association, North American Society for Cardiovascular Imaging, and Society for Cardiovascular Magnetic Resonance Imaging Expert Consensus Document on Cardiovascular Magnetic Resonance indicates that the use of CMR stress testing (vasodilator or dobutamine) is appropriate in individuals with intermediate pretest probability of CAD, patients with an uninterpretable EEG, or those unable to exercise. CMR stress testing is also appropriate for cardiac risk assessment in patients with prior coronary angiography or stenosis of unclear clinical significance.


Clinical Validation of MR Perfusion Imaging


Many single-center studies have demonstrated excellent performance of CMR for the detection of CAD compared with other imaging modalities. The few multicenter studies that have been performed, which probably better reflect the real world than specialized single-center studies, have similarly demonstrated the superiority of CMR over other imaging techniques. In a large prospective trial, including 752 patients, in which CMR and SPECT were compared with x-ray coronary angiography as the reference standard, multiparametric CMR (stress perfusion, rest perfusion, and late gadolinium enhancement) had better sensitivity than SPECT (86.5% vs 66.5%) and NPV (90.5% vs 79.1%) for diagnosis of CAD. When considered alone, the stress perfusion part of the examination also out-performed SPECT both in the single-vessel and the multivessel disease groups. Stress myocardial perfusion MR also helps with risk stratification of patients with known coronary artery stenosis of intermediate angiographic severity and may help identification of patients at risk of major adverse cardiovascular event (ie, death, stroke, and MI). Patients with coronary lesions of intermediate severity (50%–75% diameter stenosis) with myocardial perfusion defect have significantly higher incidence of adverse outcome (4%–20%) than those with equivalent coronary artery stenosis without perfusion abnormality. In addition, a normal stress test is associated with low event rate. In a series of 513 patients with known or suspected CAD, those with normal CMR stress test (adenosine stress perfusion and dobutamine stress wall motion) had a 3-year event-free survival rate greater than 99%.


Two different recent meta-analyses have reported CMR perfusion imaging to have sensitivity of 89% to 91% and specificity of 80% to 81% for the detection of CAD. The exact role of myocardial perfusion MR imaging in patients with acute chest pain and acute coronary syndrome presenting to the emergency department is still to be defined. Some studies have demonstrated, however, that CMR perfusion imaging is feasible in these patients and has high diagnostic accuracy for detection of true acute coronary syndromes, particularly when adenosine vasodilator stress is used (sensitivity: 77%–100%, specificity: 83%–93%, and accuracy: 87%). Follow-up after normal perfusion CMR examinations has demonstrated an excellent NPV of 94% to 100% for adverse outcome or subsequent diagnosis of CAD after hospital discharge. An imaging strategy involving perfusion CMR may also reduce cost in patient care by reducing unnecessary admissions and cardiac catheterizations without missing true positive acute coronary syndromes. It remains unclear how versatile and cost-effective CMR would be compared with other imaging modalities (ie, SPECT) used for early detection of acute coronary syndrome in the emergency department.


Imaging Technique


First-pass perfusion CMR examination is based on dynamic rapid imaging of the heart during the circulation of a gadolinium-based contrast agent from the cardiac cambers into the myocardium. CMR perfusion imaging can be performed as a stand-alone technique or more commonly as part of a comprehensive CMR protocol (multiparametric examination), which includes noncontrast acquisition for morphologic and functional assessment as well as delayed enhancement (DE) images. Different protocols exist for comprehensive examination of the heart, but all contain morphologic and functional sequences; a stress dynamic first-pass perfusion after contrast administration, which is compared with a similar acquisition at rest; and finally delayed images for the detection of late gadolinium enhancement. Many variations exist depending on the type of magnet, coils, clinical question at hand, stressor agent used, and personal preferences.


A typical MR protocol begins with localizers that are used to determine the true left ventricular short and long axes. These localizers are usually obtained as single-shot technique either as half-Fourier acquisition single-shot turbo spin-echo (HASTE) or as steady-state free precession (SSFP). Subsequently, functional images with white blood techniques are obtained in the short axis from the mitral valve through the apex, in vertical long axis, and in horizontal long axis views. Next, perfusion MR is obtained during the first pass of a gadolinium-based contrast agent after intravenous injection during pharmacologic vasodilation with adenosine or dipyridamole. Approximately 10 to 15 minutes later, allowing for contrast media elimination from the circulation, rest imaging is performed in the same plane and with an identical sequence (short-axis saturation recovery). Finally myocardial viability and infarction are evaluated with DE technique in which a heavily T1-weighted segmented gradient-recalled echo (GRE) sequence is acquired in at least 2 planes, usually short axis and vertical long axis. Some investigators prefer to perform the stress myocardial perfusion study first, followed by MR coronary angiography and resting myocardial perfusion study. Subsequently, additional intravenous injection of gadolinium is administered in preparation for delayed images. In the meantime, resting wall images are obtained, with DE images acquired at the end. Other investigators prefer to perform the resting perfusion examination first, followed by cine images, MR angiography, and subsequently the stress perfusion and viability examination performed at the end of the exam. A stress-only protocol has been proposed based on the high diagnostic performance of the hyperemia data from different studies. Adding T2-weighted imaging for depiction of edema has also been proposed for better detection of patients with acute coronary syndrome. The order in which stress and rest perfusion are performed may also be influenced by whether adenosine or dipyridamole is used as the stressor agent. Because adenosine has a short half-life (<10 seconds), it may be better to perform the stress study first so that the rest perfusion examination is not influenced by residual gadolinium injected at rest, which may accumulate in areas of scar tissue. If dipyridamole is used, it may be better to perform the rest study first, and then the stress study, because it has a significant longer half-life (30 minutes) and may delay the completion of the study ( Fig. 2 ).




Fig. 2


( A ) Typical first-pass perfusion protocol for adenosine stress perfusion CMR. ( B ) Typical first-pass perfusion protocol for dipyridamole stress perfusion CMR. ( C ) Typical first-pass perfusion protocol for regadenoson stress perfusion CMR.

( Modified from Gerber BL, Raman SV, Nayak K, et al. Myocardial first-pass perfusion cardiovascular magnetic resonance: history, theory, and current state of the art. J Cardiovasc Magn Reson 2008;10:18.)


Stress Agents


Adenosine is a naturally occurring substance that activates A 2 cell-surface receptors of vascular smooth muscle cells causing relaxation and hence vasodilation. Unlike normal vessels or mildly abnormal arteries, the more diseased coronary arteries have a reduced blood flow reserve and cannot further dilate in response to adenosine, which creates a heterogeneous flow pattern across the myocardium. The vasodilatory effect may create mild to moderate reduction in systolic and diastolic blood pressure and reflex tachycardia. Significant advantages of adenosine over other stress agents used in cardiovascular imaging are its short half-life and excellent safety profile, with rare serious complications. Contraindications of adenosine include second-degree and third-degree atrioventricular block, sick sinus syndrome, symptomatic bradycardia, severe asthma, and chronic obstructive pulmonary disease.


Dipyridamole is an indirect coronary vasodilator that inhibits intracellular reuptake and deamination of adenosine, increasing intravascular adenosine levels. Regadenoson, a selective A 2A adenosine receptor agonist, is a newly introduced stressor agent that induces functional and perfusion results similar to nonselective adenosine and is used for stress perfusion imaging examinations. Regadenoson has several advantages over adenosine: it is administered as an intravenous bolus fixed dose (400 μg independent of the patient weight); has fewer adverse effects, including atrioventricular block and bronchospasm; can be used in patients with mild to moderate reactive airway disease; and is cheaper. Patients should be advised to restrain from smoking, drinking tea, coffee, or any caffeine-containing substance at least for 12 hours (some prefer up to 48 hours) previous to the examination due to the competitive interaction between caffeine and adenosine or regadenoson, which may attenuate the coronary hyperemic response by competitive blockade of A 2A receptors. For stress perfusion imaging, intravenous adenosine is administered at a dose of 140 μg/kg/min. Approximately 4 minutes into the adenosine infusion, an intravenous bolus injection of 0.05 mmol/kg to 0.01 mmol/kg of gadolinium-based contrast at 3 mL/min to 4 mL/min is administered, followed by 15 mL of saline flush at 5 mL/s, during end-expiratory breath hold.


The majority of clinical trials published so far have been done with 1.5-T magnets but the 3-T magnets are being used more and more in the clinical arena. Imaging at 3 T differs significantly from 1.5 T because it suffers from increased susceptibility artifacts, difference in tissue relaxation, and radiofrequency homogeneity problems. At the same time, higher field strength increases the signal-to-noise ratio, which theoretically improves image quality and may reduce imaging time. Only a few studies of myocardial perfusion performed by 3-T magnets have been published, but so far they tend to demonstrate a better contrast and spatial resolution, with improved accuracy in diagnosis of myocardial perfusion defects, with sensitivity as high as of 90%.


Image Interpretation


Qualitative interpretation of perfusion CMR imaging is the most common approach in clinical practice. Image interpretation usually begins with review of delayed images for the presence of scar from previous MI. In absence of scar, the nonischemic myocardium exhibits uniform enhancement during first-pass perfusion at rest and with vasodilator stress. Ischemic but noninfarcted myocardium shows perfusion abnormality at stress that normalizes at rest ( Fig. 3 ). Matched stress and rest perfusion defects in the absence of DE in the same region are considered artifactual. Quantitative analysis of CMR perfusion examination is possible but complex. It can be significantly affected by several artifacts and variables (timing of saturation pulse, concentration of the contrast agent, magnetic field strength, phase of the cardiac cycle, and so forth) and has not been entirely validated or standardized. Therefore, quantitative analysis is more commonly used in research than in clinical practice.




Fig. 3


Stress perfusion CMR. ( A ) Stress image shows subendocardial hypoperfusion of the interventricular septum ( arrow ). ( B ) Corresponding rest image reveals normal enhancement of the ischemic zone ( arrow ).


A common artifact that is seen in as many as 1 in every 4 studies is the endocardial dark rim artifact, which can be seen in dynamic sequences when the bolus of gadolinium first enters into the left ventricle and mimics a perfusion defect ( Fig. 4 ). This dark rim is considered an artifact from susceptibility effect from gadolinium itself, from motion, or from resolution effect, because the artifact is perpendicular to the direction with the lowest spatial resolution (typically the phase-encoding direction). The most common location is at the interventricular septum in the left anterior descending artery vascular territory between the right and left ventricular cavities (>50%). Unlike true perfusion defects, this artifact is transient and does not remain after the bolus of contrast, a feature that helps in their differentiation.




Fig. 4


Dark rim artifact versus perfusion defect. ( A ) The dark rim artifact is seen as a thin endocardial low signal region ( white arrows ). ( B ) Perfusion defect in the interventricular septum ( black arrow ). The dark rim artifact is usually transient whereas real perfusion defect persist longer during the first pass of the bolus of contrast.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Sep 18, 2017 | Posted by in MAGNETIC RESONANCE IMAGING | Comments Off on Contrast-Enhanced Cardiac Magnetic Resonance Imaging

Full access? Get Clinical Tree

Get Clinical Tree app for offline access