Version May 2024
INTRODUCTION — This topic will discuss clinical applications of cardiovascular magnetic resonance imaging (CMR), which is magnetic resonance imaging (MRI) of the heart and blood vessels.
Other modalities of cardiovascular imaging are discussed in separate topic reviews, including echocardiography, computed tomography, and nuclear imaging. (See "Transthoracic echocardiography: Normal cardiac anatomy and tomographic views" and "Overview of stress radionuclide myocardial perfusion imaging" and "Cardiac imaging with computed tomography and magnetic resonance in the adult" and "Coronary artery calcium scoring (CAC): Overview and clinical utilization" and "Tests to evaluate left ventricular systolic function" and "Approach to evaluation of the right ventricle in adults".)
CMR findings in individuals with coronavirus disease 2019 (COVID-19) are discussed separately. (See "COVID-19: Cardiac manifestations in adults", section on 'Cardiovascular magnetic resonance'.)
TECHNIQUES — Among the many techniques employed on MRI systems, three are the mainstays of clinical CMR [1].
●Spin echo imaging – Spin echo imaging depicts the tissue structures of the heart as bright and the blood pool as dark (the black blood approach). The spin echo method is predominantly used for anatomical imaging, and for identifying the fatty infiltration of the right ventricular free wall frequently seen in arrhythmogenic right ventricular cardiomyopathy (ARVC) [2].
●Steady state free precession imaging – Steady state free precession imaging generates images in which the blood pool appears bright and the myocardium dark similar to the older gradient echo approach. This approach is used to evaluate left and right ventricular cavity sizes and function, ventricular mass, intracardiac shunts, valvular functions, and to detect intracardiac masses. Steady state free precession can generate high temporal (less than 30 msec) and spatial (2 mm in-plane) resolution cine images in an 8- to 12-second breathhold.
●Flow velocity encoding – Flow velocity encoding (also known as phase contrast) is a technique that is used to directly measure blood flow and is useful for quantifying the severity of valvular regurgitation and stenosis, intracardiac shunt size, and the severity of arterial vascular stenosis [3].
●Other – Precise assessment of myocardial dynamics can be performed using radiofrequency (RF) tagging [4]. There is a general trend to develop new methods to perform CMR with less use of gadolinium contrast agents. Magnetic resonance spectroscopy has been used as a means of evaluating myocardial metabolism. The most widely employed approach uses phosphorus-31 (naturally abundant myocardial phosphates) as a means to evaluate the changes in the high energy phosphates (including adenosine triphosphate [ATP] and phosphocreatine [PCr], using the PCr/ATP ratio) as the most direct means for detecting myocardial ischemia [5]. The resonance position of the inorganic phosphate can be used to determine the intracellular pH.
●Newer technologies – MRI and positron emission tomography (PET) have been combined into one system. This system has the best technologies for both PET (using time of flight [TOF] method to optimize the imaging studies) and 3T MRI. In addition to providing both PET and MRI images superimposed on one another, the strengths of each technology are used (ie, improved myocardial perfusion and viability imaging with PET and high resolution imaging of cardiac structure and function with high field 3T CMR).
●Newer magnet technologies allow imaging of the heart at field strengths substantially higher than the more commonly used field strength of 1.5 Tesla. Many laboratories use 3-Tesla magnets to shorten acquisition time, improve resolution, and even 4- and 7-Tesla magnets, which are more likely to allow improved phosphorus-31 spectroscopy [6,7]. The latter two field strengths are presently used in clinical research laboratories.
GATING — Although real time CMR methods (acquisition of an entire image in less than 100 msec) are available and sometimes used, this approach suffers from lower temporal and spatial resolution. Thus, cardiac gating is generally used for CMR since data are typically acquired during many cardiac cycles to optimize spatial resolution. Robust ECG gating generally enables good spin-echo and cine image quality during sinus rhythm and even during atrial fibrillation or in the presence of occasional atrial or ventricular premature beats [8]. Although flow velocity encoding imaging can be performed in the presence of atrial fibrillation, image quality may be degraded.
Although most CMR imaging can be performed during breath holds, longer image acquisitions (eg, for high resolution coronary artery imaging) require respiratory gating in addition to cardiac gating. Respiratory gating can be accomplished using either a navigator approach (to track the motion of the diaphragm) or respiratory bellows (using an elastic band wrapped around the thorax) to monitor respiratory motion.
SAFETY — CMR has no ionizing effects on patients or caregivers. However, CMR is potentially problematic in patients with ferromagnetic metallic implants. Thorough screening of all patients prior to CMR imaging or spectroscopy is mandatory. Devices such as prosthetic heart valves, prosthetic joints, sternal wires, and intravascular stents do not preclude study with CMRI at field strengths of 1.5 and 3.0 Tesla. Issues related to the safety of MRI, including specific recommendations regarding cardiovascular devices (eg, pacemakers), are discussed separately. (See "Diagnostic imaging in pregnant and lactating patients" and "Patient evaluation for metallic or electrical implants, devices, or foreign bodies before magnetic resonance imaging", section on 'Assessing implants, devices, or foreign bodies for MRI'.)
Kidney disease — Gadolinium contrast agents administered to patients with moderate to severe kidney disease (estimated glomerular filtration rate (eGFR) less than 30 mL/min) has been rarely associated with the syndrome of nephrogenic systemic fibrosis. It is recommended that gadolinium-based imaging be avoided in such patients. This issue, including the definition of patients at risk, is discussed separately. (See "Nephrogenic systemic fibrosis/nephrogenic fibrosing dermopathy in advanced kidney disease" and 'Gadolinium contrast techniques' below.)
CLINICAL APPLICATIONS — CMR uses high magnetic fields and radiofrequency to generate 3D/tomographic images with high spatial resolution and excellent contrast. In 2002, an American Heart Association scientific statement made recommendations for standardized myocardial segmentation and nomenclature used with tomographic imaging of the heart (figure 1A-C) [9]. CMR has been used for a growing list of applications [10,11]. The advantages and disadvantages of CMR are summarized in a table (table 1).
AORTIC DISEASE — The diagnosis and assessment of diseases of the aorta is an area in which CMR excels [12]. Spin-echo and cine steady state free precession (SSFP) CMR can precisely define the dimensions and extent of complex aortic aneurysms, false aneurysms, dissection flaps, periaortic abscesses (eg, with endocarditis), aortic arch abnormalities, and coarctation. It can also be used to monitor patients with thoracic and abdominal aortic aneurysms and patients at risk for developing aortic aneurysms or dissection. In patients with homozygous familial hypercholesterolemia, CMR can detect atherosclerotic plaque and supravalvular aortic stenosis [13].
CMR is used for the diagnosis, characterization, and determination of the extent of aortic dissection (image 1). It can provide information about the differential flow velocity in the true and false lumens, and can identify involvement of branch arteries anywhere from coronaries to the femoral arteries. Because of the inherent 3D nature of the technique, the images of the aorta and its pathology can be readily reconstructed into 3D display of the aorta and its branches, thereby facilitating planning for surgical repair. (See "Clinical features and diagnosis of acute aortic dissection".)
Hemodynamic instability limits the utility of CMR in some patients with aortic dissection. Time of acquisition (in older systems), difficulty with access to the patient, while in the magnet bore, and distortion of the electrocardiogram by the magnetic field make monitoring more difficult. CMR systems optimized for cardiac and vascular imaging can acquire 3D images of the entire thoracic aorta within 25 seconds.
PERICARDIAL DISEASE — CMR permits direct visualization of the normal pericardium, which is composed of fibrous tissue and has a low CMR signal intensity with spin echo imaging [14]. The normal pericardium measures less than 3 mm in thickness. Although pericardial fluid also has a low signal intensity, it can be distinguished from pericardium with cine SSFP CMR images on which transudative pericardial fluid has a very bright signal in contrast to the dark line of the pericardium.
CMR is the diagnostic procedure of choice for the detection of certain pericardial diseases, such as noncalcified constrictive pericarditis, tumor invasion of the pericardium, and congenital absence of the pericardium [14-19]. Myocardial tagging techniques can demonstrate tethering of pericardium to the heart during the cardiac cycle. Furthermore, real-time cine imaging allows direct visualization of the effects of respiration on ventricular filling and septal configuration to assist in the diagnosis of pericardial constriction [20]. In addition, the presence of pericardial effusion can be established and the distinction made between hemorrhagic and nonhemorrhagic effusion [21,22]; however, echocardiography is the first-line approach. (See "Constrictive pericarditis: Diagnostic evaluation" and "Pericardial effusion: Approach to diagnosis".)
Pericardial enhancement seen after administration of intravenous gadolinium contrast agent suggests an inflammatory process [23]. Furthermore, tissue characterization can give important diagnostic information in patients with primary and metastatic cardiac tumors [24,25].
MYOCARDIAL DISEASE — CMR enables characterization of the range of myocardial diseases from ischemic to inflammatory and various types of cardiomyopathy. It enables visualization and quantification of the size, mass, and global and regional function of the left and right ventricular cavities. Myocardial scar and fibrosis can be identified by gadolinium contrast enhancement techniques. (See "Tests to evaluate left ventricular systolic function".)
Gadolinium contrast techniques
Late gadolinium enhancement — Late gadolinium enhancement (LGE) is thought to reflect fibrosis and irreversibly damaged myocardium including acute and chronic myocardial infarction (MI).
The possible role of LGE as a predictor for risk of subsequent cardiac events was evaluated in a group of 195 patients without known prior MI, but with known or suspected coronary artery disease [26]. During an average follow-up of 16 months, patients with LGE on the baseline study had a higher incidence of major adverse cardiovascular endpoints (MACE) compared with those without this finding. In addition, LGE was the strongest predictor of MACE in contrast to traditional clinical, angiographic, and functional variables.
The proper role of this diagnostic tool in the evaluation of patients with suspected or known coronary artery disease remains to be determined.
If LGE CMR is being considered as a diagnostic option, renal function should be assessed. (See 'Kidney disease' above and "Nephrogenic systemic fibrosis/nephrogenic fibrosing dermopathy in advanced kidney disease".)
Infarct detection and sizing — Acute and chronic MI can be demonstrated with LGE. This technique may be used to identify the location and extent of infarction [27-30]. (See "Suboptimal reperfusion after primary percutaneous coronary intervention in acute ST-elevation myocardial infarction".)
Infarct sizing is precise and highly reproducible, with minimal interobserver and intraobserver variability [29]. Imaging is typically performed 10 to 20 minutes after intravenous injection of 0.1 to 0.2mmol/kg of gadolinium chelate [1].
A multicenter, double-blinded, randomized trial evaluated the efficacy of infarct detection by LGE in 282 patients with acute MI (up to 16 days post MI) and 284 patients with chronic MI (17 days to 6 months post MI) using varying doses of gadoversetamide (from 0.05 to 0.3 mmol/kg) [30]. Comparing the doses of 0.1, 0.2, and 0.3 mmol/kg at 10 minutes after contrast administration, the sensitivity of CMR for detecting MI increased with a dose of gadoversetamide from 84 to 95 to 99 percent for acute MI and from 83 to 87 to 94 percent for chronic MI, respectively. The accuracy for identifying MI location (compared with infarct-related artery perfusion territory) also increased with a dose of contrast agent from 78 to 93 to 99 percent for acute MI and from 80 to 89 to 91 percent for chronic MI for doses of 0.1, 0.2, and 0.3 mmol/kg, respectively.
In a study in both animals and humans, LGE imaging was superior to single photon emission computed tomography (SPECT) for infarct detection in patients with known or suspected CAD [27]. CMR identified 181 segments with subendocardial MI in 91 patients with known or suspected coronary disease compared with only 96 segments detected by SPECT. SPECT was more comparable to CMR in detecting transmural MI. The superiority of CMR for subendocardial MI detection is likely due to the greater spatial resolution of CMR.
In a direct comparison of CMR and SPECT in the setting of acute MI, 78 patients who underwent urgent percutaneous coronary intervention were evaluated by both techniques seven days after the infarct [31]. CMR was significantly more sensitive than SPECT for all infarcts (97 versus 87 percent), small infarcts (92 versus 69 percent), and nonanterior infarcts (98 versus 84 percent) and there was a nonsignificant trend toward increased sensitivity for non-Q wave infarction (85 versus 46 percent).
CMR may also detect the presence of microvascular obstruction, which leads to severe reduction in blood flow; this leads to a central dark area on inversion recovery with CMR images performed one to two minutes (early) after contrast injection [32-36]. The extent of microvascular obstruction and infarct size increases substantially over the first 48 hours after an MI [32].
The clinical significance of CMR-defined microvascular obstruction after an MI has been assessed in small clinical studies. In a report of 44 patients, those with CMR-defined microvascular obstruction had a significantly higher rate of cardiovascular events during a 16-month follow-up (45 versus 9 percent for those without this abnormality) [33].
In addition, LGE in the infarct zone within 24 hours after primary PCI is an independent predictor of impaired left ventricular systolic thickening and remodeling [34,35].
In the setting of inferior wall MI, LGE has been compared with physical examination, electrocardiography, and echocardiography for the detection of right ventricular infarction [37]. In an initial report, LGE CMR was significantly more sensitive than any of the other methods. The clinical and prognostic relevance of this finding remains to be determined. (See "Right ventricular myocardial infarction", section on 'Diagnosis'.)
Myocardial viability — LGE can also be used to assess myocardial viability (actually nonviability or MI). Studies in laboratory animals have found that, independent of wall motion or infarct age, regions exhibiting gadolinium contrast enhancement at least 10 minutes after the infusion of gadolinium-based contrast agents coincide with regions of myocardial necrosis and irreversible myocardial injury; regions that fail to enhance are viable [38,39].
Clinical studies have confirmed that a normal LGE pattern occurs in dysfunctional myocardium that is viable and displays improved contractile function in response to low dose (5 to 10 mcg/kg/min) dobutamine infusion, while central regions with enhancement where the infarct is transmural display no contractile activity in response to the dobutamine infusion. Territories that have nontransmural necrosis display a diminished contractile response to dobutamine [40]. LGE as a marker of scar closely agrees with the finding of matching defects on PET viability scanning [41]. (See "Assessment of myocardial viability by nuclear imaging in coronary heart disease", section on 'Positron emission tomography'.)
Further support for these findings comes from a clinical study of 32 patients with a proven MI who underwent coronary angiography; LGE, performed 3 or 14 months after the MI, accurately established the presence, location, and transmural extent of healed Q wave and non-Q wave MI [42]. Large infarcts were predominantly transmural, while small infarcts were nontransmural (image 2). The transmural extent of infarction predicts improvement in left ventricular function. In one study of 24 patients, the extent of dysfunctional myocardium that was not infarcted or had necrosis comprising <25 percent of left ventricular wall thickness, as established by LGE performed within one week of the MI, was the best predictor of global improvement in contractility at three months [43].
A reduction in the extent of LGE early after an MI is associated with early restoration of blood flow and improvement in contractile function [44]. This is in contrast to what has been seen with microvascular obstruction and early hypoenhancement immediately after an infarction, as discussed above; this myocardium has no contractile response to dobutamine [40].
LGE CMR can also be used to establish the presence of hibernating myocardium. In one series of 24 patients with stable coronary artery disease and left ventricular dysfunction, LGE (three to 15 minutes after contrast administration) was associated with nonviability as established with rest-distribution thallium imaging and dobutamine echocardiography; the absence of enhancement correlated with radionuclide and echocardiographic viability, regardless of the status of resting contractile function (image 3) [45]. (See "Evaluation of hibernating myocardium" and "Dobutamine stress echocardiography in the evaluation of hibernating myocardium" and "Assessment of myocardial viability by nuclear imaging in coronary heart disease".)
The extent of enhancement with LGE can predict recovery of left ventricular systolic function after revascularization [46,47]. As an example, one study of 50 patients with coronary artery disease who had left ventricular dysfunction prior to surgical or percutaneous revascularization found that 33 percent of myocardial segments in 80 percent of patients had evidence of LGE; 38 percent of segments had abnormal contractility [46]. After revascularization, more dysfunctional segments without LGE improved (78 versus 17 percent with enhancement of more than 75 percent of the tissue). The likelihood of improvement in regional contractility after revascularization decreased progressively as the transmural extent of LGE increased. The percentage of the left ventricle that was dysfunctional and not enhanced was significantly related to the degree of improvement in left ventricular ejection fraction. (See "Treatment of ischemic cardiomyopathy".)
Because the size of LGE enhancement may decrease with time, there may be predictive value in assessing nonenhanced regions of the ventricle. One study demonstrated the value of measuring the nonenhancing wall thickness to predict improvement in systolic wall thickening [48].
CMR myocardial tagging is another noninvasive method that quantifies local myocardial segment shortening, including transmural wall motion [49]. Quantitative analysis of regional intramural function can be performed independent of accurate border detection when this technique is used in conjunction with two-dimensional strain analysis [50].
The term "strain" refers to an object's fractional or percent change from its original, unstressed, dimension (ie, a change in length corrected for the original length) [51]. It reflects deformation of a structure, and when applied to the myocardium, strain directly describes the contraction/relaxation pattern. Strain rate is the rate of this deformation [51].
When combined with low-dose dobutamine, CMR myocardial tagging can quantify the amount of myocardial viability after an acute MI and provide prognostic information. In one study of 20 patients with a first reperfused MI, a normal increase in shortening within the midwall and subepicardium, but not the subendocardium, during a low-dose dobutamine infusion predicted greater functional recovery eight weeks after the MI [52]. A second report compared dobutamine CMR with thallium and tetrofosmin SPECT imaging in 30 patients [53]. The sensitivity and specificity of dobutamine CMR were 50 and 81 percent, respectively; SPECT imaging with thallium or technetium-99m tetrofosmin had higher sensitivity (76 and 66 percent, respectively) but lower specificity (44 and 49 percent, respectively).
Prediction of post-MI mortality — CMR has been used to evaluate the size of the peri-infarct border zone size in relation to the core infarct as a predictor of post-MI mortality [54]. The peri-infarct border zone is postulated to represent a mixture of viable and scarred myocardium, a heterogeneous region that may represent substrate for ventricular arrhythmias. Thus, the size of the peri-infarct border zone may predict future arrhythmic risk. (See "Risk stratification after acute ST-elevation myocardial infarction" and "Risk stratification after non-ST elevation acute coronary syndrome".)
The survival of 144 patients with documented coronary artery disease and LGE consistent with infarction was assessed at a median follow-up of 2.4 years [54]. Patients with a peri-infarct zone to total infarct (core and peri-infarct zone) ratio above the median were at higher risk of death (28 versus 13 percent in those below the median). This emerging methodology requires validation.
Chest pain without significant coronary artery disease — In various clinical trials and the CRUSADE registry, 9 to 14 percent of patients with a non-ST elevation acute coronary syndrome (ACS) have, on coronary angiography, either minimal disease or no vessel with at least 50 to 60 percent stenosis. In addition, as many as 7.5 percent of patients with an ST-elevation MI do not have a critical coronary artery lesion and approximately 3 percent have normal epicardial coronary arteries. (See "Acute coronary syndrome: Terminology and classification", section on 'Absence of significant coronary disease'.)
The possible role of LGE in the diagnostic evaluation of troponin positive chest pain patients without coronary obstruction was investigated in 60 such patients, 40 percent of whom had ST elevation on the initial ECG [55]. CMR was performed within three months (median interval of 15 days) of initial presentation. Based on the LGE pattern, a cause for the troponin elevation was identified in 65 percent of the patients, with myocarditis in 50 percent, MI in 12 percent, and cardiomyopathy in 3 percent.
CMR has also been used to demonstrate abnormal subendocardial perfusion in patients with cardiac syndrome X with chest pain and normal epicardial coronary arteries [56]. (See "Microvascular angina: Angina pectoris with normal coronary arteries".)
Detection of diffuse myocardial fibrosis — LGE is useful for detection of focal but not diffuse fibrosis. An equilibrium contrast CMR (EQ-CMR) technique has been proposed as a means to identify diffuse myocardial fibrosis [57]. However, there are not enough data to make any definitive comment on this approach. The technique was clinically tested in a small population of patients by demonstration of a good correlation between EQ-CMR and histologic fibrosis (r2 = 0.80) in 18 patients with aortic stenosis and eight patients with hypertrophic cardiomyopathy.
T1 mapping — T1 mapping is a technique that calculates a numerical value for the T1 relaxation of the myocardium. This technique has been used to identify abnormalities in various disorders, including myocarditis, cardiac amyloidosis, hypertrophic cardiomyopathy, iron overload states, and Fabry disease [58-61]. Native T1 values over 1136 ms are strongly suggestive of cardiac amyloid in patients with a clinical suspicion [62]. Despite the lack of standardization of the acquisitions and some overlap between normal and disease states, T1 mapping and ECV are increasingly being applied to clinical practice.
CMR characterization of myocardial diseases
Ischemic heart disease — CMR can be useful in assessing patients with possible or definite ischemic heart disease. Since CMR provides a dimensionally accurate, 3D perspective of the heart, it has been considered by many to be the "gold standard" for measurements of left and right ventricular ejection fraction, volumes, and myocardial mass. It can detect regional wall motion abnormalities and high-energy phosphate abnormalities in ischemic tissue [63].
Acute coronary syndrome — The role of CMR for the diagnostic evaluation of acute coronary syndrome (ACS) is under investigation.
●An early report evaluated the use of CMR in 161 patients presenting to an emergency department with chest pain suggestive of an ACS and a nondiagnostic ECG [64]. CMR was obtained within 12 hours and included assessment for regional wall motion abnormalities, myocardial perfusion defects, and MI using LGE. Sensitivity and specificity of CMR for diagnosing ACS were 84 and 85 percent, respectively. Further analysis showed that CMR was the strongest predictor of ACS and provided incremental diagnostic information over clinical parameters such as ischemic ECG changes, peak troponin-I, and TIMI risk score.
●Similar findings have been reported in a single-center trial randomizing 105 intermediate-risk patients with suspected ACS to either usual care or care in an observation unit with stress CMR imaging, in which the patients who underwent stress CMR were noted to have significant reductions in length of stay, coronary revascularization, hospital readmission, and recurrent cardiac imaging [65]. While encouraging, these findings will need to be replicated in larger multicenter studies prior to the routine adoption of stress CMR for evaluation of patients with suspected ACS.
●An augmented CMR protocol (including T2-weighted imaging and assessment of left ventricular wall thickness) was compared with a standard CMR protocol (function, perfusion, and LGE) in 62 patients presenting to the emergency department with acute chest pain with negative cardiac biomarkers and no acute ischemic ECG changes (see 'Chest pain without significant coronary artery disease' above). As compared with the standard protocol, specificity, positive predictive value, and overall accuracy increased from 84 to 96 percent, 55 to 85 percent, and 84 to 93 percent respectively when using this novel protocol. In addition, the augmented CMR protocol demonstrated value over that of the initial clinical risk assessment and traditional risk factors [66].
Pharmacologic stress CMR — For detection of significant CAD, CMR is used in conjunction with pharmacologic stress. The main pharmacologic agents are dobutamine for stress testing and vasodilator agents, including regadenoson, adenosine, and dipyridamole.
Dobutamine — Dobutamine infusion has been used to induce stress with CMR imaging, allowing visualization of reduced wall motion in segments that become ischemic [67-71]. This technique provides an alternative for the diagnosis of coronary artery disease and for quantification of "myocardium at risk"; it may be particularly useful for patients in whom adequate echocardiograms cannot be obtained [68].
One study compared dobutamine stress echocardiography with dobutamine stress CMR in 208 patients with suspected coronary disease who underwent radiograph coronary angiography. Compared with dobutamine stress echocardiography, dobutamine CMR had greater sensitivity (86 versus 74 percent) and specificity (86 versus 70 percent); the results were similar in men and women [69].
Low-dose dobutamine CMR may also be useful in predicting left ventricular functional recovery after coronary revascularization [70,71]. In a series of 43 patients with a prior MI who subsequently underwent coronary artery revascularization, dobutamine-induced systolic wall thickening predicted functional recovery of the affected wall with a sensitivity and specificity of 89 and 94 percent, respectively (image 4A-B) [70]. (See "Treatment of ischemic cardiomyopathy".)
CMR myocardial tagging is a noninvasive method that quantifies local myocardial segment shortening throughout the left ventricular myocardium at sites across the left ventricular wall thickness [4]. Myocardial tagging improves the sensitivity of dobutamine stress CMR. (See 'Myocardial viability' above.)
The potential prognostic value of CMR myocardial tagging in this setting was evaluated in 194 patients with suspected myocardial ischemia [72]. Dobutamine stress CMR with tagging detected more new regional wall motion abnormalities with stress than dobutamine stress CMR without tagging (68 versus 58 patients). Cardiac catheterization demonstrated coronary artery disease in 65 of the 68 patients with a positive tagged scan; 62 of these required revascularization. In contrast, among the patients with a negative tagged scan, the event-free survival rate was 98.2 percent at 17 months.
Dipyridamole — Ischemic territories can be identified by using intravenous dipyridamole and gadolinium for a first-pass perfusion study and estimation of a myocardial perfusion reserve index [73,74]. As an example, one study of 48 patients and 18 healthy subjects found that a coronary flow reserve less than 1.65 had a sensitivity and specificity of 91 and 94 percent, respectively, for the detection of coronary artery disease as defined by positron emission tomography (PET) scanning; sensitivity and specificity were 87 and 85 percent, respectively, when compared with quantitative coronary radiograph angiography [74]. This technique was accurate even when perfusion abnormalities were confined to the subendocardium.
The myocardial perfusion reserve index, obtained from CMR is also useful for assessing the results of a percutaneous coronary intervention; one study found that this index improved, but did not normalize, after angioplasty, while it completely normalized after angioplasty with stenting [75].
Adenosine — Adenosine stress CMR has been used to detect myocardial ischemia by inducing wall motion abnormalities and by creating relative perfusion differences seen in first pass perfusion imaging [76-78]. In a report comparing CMR adenosine-induced wall motion abnormalities and dobutamine-induced wall motion abnormalities, the sensitivity of adenosine testing for significant stenosis as determined by quantitative angiography was much lower with adenosine wall motion stress (40 versus 89 percent), while the specificity was higher (96 versus 80 percent) [77].
The diagnostic accuracy of adenosine stress CMR varies significantly according to the CMR parameter evaluated. Direct assessment of myocardial perfusion after adenosine stress is sensitive, while adenosine-induced wall motion abnormalities have a high specificity [77]. The performance of adenosine stress CMR may be improved by the combination of stress perfusion imaging and LGE infarction imaging [79]. In the CE-MARC trial, comprehensive CMR including adenosine perfusion was compared with standard SPECT for the detection of significant stenosis with radiograph angiography as the gold standard in 752 patients. The sensitivity of CMR was found to be superior to that of SPECT with similar specificity [80].
Regadenoson — Regadenoson is now commonly used to induce vasodilation and is safer and more effective than adenosine and more effective than dipyridamole. It is the most common agent now employed for PET myocardial perfusion imaging. While it is more expensive, it works rapidly with fewer side effects.
Adenosine plus dobutamine — In a report assessing the prognostic value of CMR using combined, single session adenosine and dobutamine stress, 513 patients with known or suspected coronary disease were evaluated during a mean follow-up of 2.3 years [81]. The three-year event-free (cardiac death and nonfatal MI) survival was 99 percent with normal and 84 percent with abnormal combined stress images.
CMR after myocardial infarction — Left ventricular ejection fraction and end-systolic volume index are important prognostic indicators after acute MI. (See "Incidence of and risk stratification for sudden cardiac death after myocardial infarction".)
Acquisition of 3D stacks of CMR images permits calculation of left and right ventricular volumes without geometric assumptions which may lead to inaccuracies (especially in focally deformed ventricles) [82]. In addition, CMR can detect and quantify the many potential complications of MI. These include:
●Mitral regurgitation and ventricular septal defect. (See "Acute myocardial infarction: Mechanical complications".)
●Left ventricular thrombus. (See 'Intracavitary thrombus' below and "Left ventricular thrombus after acute myocardial infarction".)
●Left ventricular aneurysm and pseudoaneurysm. (See "Left ventricular aneurysm and pseudoaneurysm following acute myocardial infarction".)
●Pericardial effusion. (See "Pericardial complications of myocardial infarction".)
In addition, acute and chronic MI can be identified with LGE CMR. (See 'Infarct detection and sizing' above.)
Coronary MR angiography — The production of diagnostic quality angiograms of the coronary arteries with CMR is another area of rapid development. The utility of coronary MRI in patients with native vessel or bypass graft disease is discussed in detail separately. (See "Cardiac imaging with computed tomography and magnetic resonance in the adult".)
Coronary MRA may be particularly useful for the identification or characterization of anomalous coronary arteries and for monitoring coronary artery aneurysms. (See "Congenital and pediatric coronary artery abnormalities".)
Cardiomyopathy — The high spatial resolution of CMR enables accurate assessment of ventricular volumes, ventricular systolic function (ejection fraction), and myocardial mass and wall thickness. Such analysis is useful in the assessment of patients with heart failure, for the diagnostic evaluation of cardiomyopathy, for prediction of outcomes, and may frequently be the preferred diagnostic test. (See "Tests to evaluate left ventricular systolic function".)
Ischemic versus nonischemic cardiomyopathy — High-resolution evaluation of regional ventricular systolic function can help differentiate between ischemic and nonischemic cardiomyopathy. LGE, which identifies myocardial scar/fibrosis, can also be used to make this distinction.
LGE is present in most patients with ischemic cardiomyopathy (81 to 100 percent) compared with 12 to 41 percent in patients without significant obstructive coronary disease [49,83,84]. Although LGE can be seen in ischemic and nonischemic cardiomyopathies, the patterns of LGE tend to be different in the two disorders [83-85]:
●Ischemic cardiomyopathy is characterized by subendocardial and/or transmural LGE.
●In comparison, isolated mid-wall or epicardial enhancement is strongly suggestive of a nonischemic cardiomyopathy. Mid-wall involvement in ischemic cardiomyopathy involved segments different from those showing subendocardial LGE [83].
●In two studies, LGE similar to that in ischemic cardiomyopathy was seen in 9 to 13 percent of patients with unobstructed coronary arteries [83,84]. A possible explanation for this finding is recanalization after an MI [84].
LGE also may be seen in hypertrophic cardiomyopathy, myocarditis, sarcoidosis, and infiltrative cardiomyopathies such as amyloidosis [85].
The use of LGE to predict recovery of left ventricular function after revascularization in patients with ischemic cardiomyopathy and the use of CMR for infarct detection and size and for the detection of intracardiac thrombi are discussed above, while the data evaluating the use of coronary MRI are presented separately. (See 'Myocardial viability' above and 'Infarct detection and sizing' above and 'Intracavitary thrombus' below and "Cardiac imaging with computed tomography and magnetic resonance in the adult".)
Hypertrophic cardiomyopathy — Asymmetric thickening of the interventricular septum and other patterns of left ventricular hypertrophy in hypertrophic cardiomyopathy (HCM) can be readily detected by CMR. CMR may be of particular value in the assessment of variant types of hypertrophic cardiomyopathy sometimes not detected by TTE, including apical and lateral wall hypertrophy [86,87]. It is also possible to visualize systolic anterior motion of the mitral valve and turbulence in the left ventricular outflow tract created by the dynamic obstruction in obstructive hypertrophic cardiomyopathy. (See "Hypertrophic cardiomyopathy: Clinical manifestations, diagnosis, and evaluation".)
In addition, LGE techniques may be used to identify patchy mid-wall LGE, particularly at the junctions of the interventricular septum and right ventricular free wall [85,88,89]. The extent of enhancement has been associated with progressive ventricular dilation and sudden death [89,90], and the presence of enhancement has been shown to predict increased risk even in patients at low risk for sudden death by the usual risk factors. (See "Hypertrophic cardiomyopathy: Risk stratification for sudden cardiac death", section on 'Pathogenesis of arrhythmias'.)
Cardiac sarcoidosis — Diagnosis and monitoring of cardiac sarcoidosis by CMR is discussed separately. (See "Clinical manifestations and diagnosis of cardiac sarcoidosis".)
Iron overload — There is a growing body of literature on the role of CMR in patients with suspected iron overload. In the presence of myocardial iron overload, myocardial T2* (T2 star) is reduced due to inhomogeneities that are increased with iron deposition. In one study of 106 patients with thalassemia and multiple blood transfusions, there was a significant inverse correlation between liver T2* and liver iron concentration by biopsy [91]. Increasing myocardial iron content as assessed by myocardial T2* correlated with decreasing left ventricular ejection fraction but there was no correlation between myocardial T2* and serum ferritin or liver iron content, which may relate to temporal differences in iron uptake and release between these two organs [92].
A difference in iron loading and unloading between the liver and heart was also seen in a report dealing with 66 patients (age range: 3 to 82 years) with transfusion-dependent anemias and thalassemia intermedia [93]. There was a poor correlation between hepatic and myocardial iron as assessed by T2* MRI, and a subgroup of patients was identified with increased myocardial iron without a matched degree of hepatic hemosiderosis. Left ventricular ejection fraction was insensitive for detecting elevated myocardial iron. (See "Approach to the patient with suspected iron overload", section on 'Noninvasive imaging (MRI)'.)
Inter-center reproducibility of the T2* technique was validated in a small multicenter study [94]. CMR also enables evaluation of response to treatment of iron overload. A meta-analysis of available studies in patients with transfusion-dependent thalassemia has concluded that iron chelators significantly reduced myocardial iron content by approximately 24 percent (95% CI 17-30 percent) [95]. There was no significant difference between the amount of iron reduced by deferoxamine and deferiprone. LVEF was not significantly changed by chelation, although this meta-analysis indicated that a publication bias existed for LVEF, but not for myocardial iron.
The prognostic value of cardiac T2* was evaluated in a study of 652 thalassemia major patients followed at 21 UK centers who underwent 1442 CMR scans at a single CMR center [96]. At one year follow-up there were 80 episodes of heart failure and 98 episodes of arrhythmia. The following associations with T2* were noted:
●Cardiac T2* <10 ms was associated with heart failure with a relative risk of 160 (95% CI, 39 to 653). Heart failure occurred in 47 percent of patients with cardiac T2* <6 ms with a relative risk of 270 (95% CI, 64 to 1129). Cardiac T2* was <10 ms in 98 percent of scans in patients who developed heart failure.
●Cardiac T2* <20 ms was associated with arrhythmia with a relative risk of 4.6 (95% CI, 2.7 to 8). Arrhythmia occurred in 14 percent of patients with a cardiac T2* of <6 ms. Cardiac T2* was <20 ms in 83 percent of scans of patients who developed arrhythmias.
●Cardiac T2* was a substantially better predictor of heart failure or arrhythmia than liver T2* or serum ferritin.
The availability of automated T2* CMR software may facilitate dissemination of this technique for monitoring myocardial siderosis.
Restrictive cardiomyopathy versus constrictive pericarditis — CMR is also an excellent approach to differentiate between restrictive and constrictive disease [97]. (See "Differentiating constrictive pericarditis and restrictive cardiomyopathy".)
As an example, the infiltrated myocardium in restrictive cardiomyopathy due to amyloidosis is easily detected [85,98,99]. There is homogeneous thickening of the ventricular and atrial walls, including the interatrial septum and right atrial free wall (image 2). These abnormalities are not seen in patients with idiopathic restrictive cardiomyopathy (image 3) [99]. LGE imaging often demonstrates circumferential subendocardial or diffuse transmural enhancement. Elevated native T1 values and ECV are also helpful to diagnose cardiac amyloid [62]. (See "Cardiac amyloidosis: Epidemiology, clinical manifestations, and diagnosis".)
Chagas heart disease — The utility of LGE CMR in the assessment of Chagas heart disease was assessed in a series of 51 patients of varying disease severity [100]. Myocardial fibrosis was detected in early asymptomatic stages and increased with clinical stage of the disease.
Acute myocarditis — CMR can be used to identify the presence of myocardial edema and myocyte damage in myocarditis. CMR evaluation of myocarditis is discussed separately. (See "Clinical manifestations and diagnosis of myocarditis in adults".)
CMR findings in individuals with coronavirus disease 2019 (COVID-19) are discussed separately. (See "COVID-19: Cardiac manifestations in adults", section on 'Cardiovascular magnetic resonance'.)
VALVULAR HEART DISEASE — The turbulence created by valvular stenosis and regurgitation is easily visualized on cine gradient echo CMR [82,101]. However, changes in the acquisition parameters (eg, shorter echo time, use of steady state free precession sequences) can markedly alter this effect, so that qualitative assessment of valvular lesions must be done with caution and by those familiar with the characteristics of the involved parameters. In addition, CMR determined ejection fraction and ventricular volumes can help to determine the timing of valvular surgery.
Regurgitant valve disease — Regurgitant valvular lesions produce a zone of proximal isovelocity surface area (PISA) on the side of the valve opposite from the direction of regurgitant flow with both CMR and echocardiography [102]. The amount of PISA is directly related to the amount of regurgitation.
Phase velocity mapping allows detection and quantification of the amount of aortic regurgitation.
Valvular regurgitation caused by endocarditis is well seen with gradient echo images, but vegetations are only occasionally visualized [103].
Stenotic valve disease — The value of CMR in quantifying stenotic valve disease is illustrated by the following studies:
●In an initial report, aortic valve planimetry by CMR correlated well with transesophageal echocardiography (TEE) planimetry in a series of 40 patients with aortic stenosis [104]. The mean valve area was 0.91 cm2 by CMR and 0.89 cm2 by TEE; the correlation coefficient between the two methods was 0.96. However, the correlation of CMR with valve area determined by cardiac catheterization was not as good (0.44). (See "Aortic valve area in aortic stenosis in adults".)
●In two series of 39 patients with mitral stenosis, the correlation coefficients of planimetered mitral valve areas from CMR data were 0.81 and 0.86 with Doppler pressure half-time echocardiography and 0.89 with catheterization [105,106]. CMR slightly overestimated mitral valve area compared with echocardiography (by 8 percent) and catheterization (by 5 percent) [106]. Thus, a planimetered CMR mitral valve area less than 1.65 cm2 suggests mitral stenosis defined as a catheterization value of less than 1.5 cm2. (See "Rheumatic mitral stenosis: Clinical manifestations and diagnosis".)
Myocardial fibrosis — The potential prognostic value of myocardial fibrosis detected by CMR late gadolinium enhancement (LGE) in patients with aortic stenosis is discussed separately. (See "Clinical manifestations and diagnosis of aortic stenosis in adults", section on 'Cardiovascular magnetic resonance'.)
INTRACAVITARY THROMBUS — Clinical detection of ventricular thrombi is generally performed by transthoracic echocardiography (TTE) and evaluation of atrial thrombi is generally performed by TEE [107]. (See "Left ventricular thrombus after acute myocardial infarction".)
However, CMR appears to have greater sensitivity and similarly high specificity for detection of left ventricular thrombi in the MI population [108-110]. The largest study consisted of 160 patients with a remote prior infarction who had surgical and/or pathological confirmation of the presence (48 patients [30 percent]) or absence of chronic left ventricular thrombus; all patients underwent CMR including LGE, TTE, and intraoperative TEE. CMR was significantly more sensitive (88 versus 23 and 40 percent with TTE and TEE, respectively); all three imaging modalities were highly specific (99 percent for CMR versus 96 percent with TTE and TEE). The patients with left ventricular thrombus had a higher incidence of recent embolic events (6.1 versus 0.8 percent).
High LGE sensitivity for left ventricular thrombus was also noted in an analysis of 57 patients with acute or past MI or ischemic cardiomyopathy [109]. LGE detected mural left ventricular thrombus in 12 (21 percent); only five were seen by TTE [109].
Limited data are available for CMR imaging of left ventricular thrombus in nonischemic populations and of atrial thrombus. Although atrial thrombus can be detected by contrast and noncontrast CMR techniques, diagnostic accuracy compared with TEE is limited [111,112].
PERIPHERAL ARTERY DISEASE — The ability to generate images of blood vessels without the administration of contrast medium coupled with the capability of producing high resolution 3D images gives CMR a particular advantage in assessing peripheral artery disease [12]. Carotid arteries can be imaged along their entire length, identifying stenosis, resultant turbulent flow, and the presence and possibly age of arterial thrombus [113].
Preliminary data suggest that CMR can distinguish atherosclerotic lesions with intact, thick fibrous caps from those with a lipid-rich necrotic core and intraplaque hemorrhage or with thin and disrupted caps, which have the potential for rupture [114-116]. Although the number of patients who have been studied is small, patients undergoing carotid endarterectomy who have ruptured caps on CMR appear to be more likely to have had a recent transient ischemic attack or stroke than those with thick fibrous caps [116].
CMR can also be useful in documenting the regression of aortic and carotid atherosclerotic lesions during therapy with a statin drug [117]. The iliac and femoral arteries can also be readily imaged [118]. These images permit precise planning for peripheral interventions, without exposure to iodinated contrast medium or the post-procedure recovery associated with arterial puncture.
CONGENITAL HEART DISEASE — CMR has many other clinically important applications for evaluation of patients with congenital heart disease [119]. Patients with complex congenital heart disease include many who are younger and will undergo numerous imaging tests during their lifetime. Thus, a noninvasive test that does not employ ionizing radiation is preferred. Spin-echo and SSFP CMR can be readily applied [120,121]. Phase velocity mapping techniques can accurately measure the pulmonary and systemic blood flows and establish the ratio associated with intracardiac shunts [3]. 3D SSFP CMR may allow simplified image acquisition and reformatting of a single set of images into any desired projection plane [122].
LGE can also detect areas of fibrosis in patients with repaired Tetralogy of Fallot, a possible marker of late adverse outcomes [123].
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Cardiac valve disease" and "Society guideline links: Myocarditis" and "Society guideline links: Multimodality cardiovascular imaging appropriate use criteria".)
SUMMARY
●The most useful imaging techniques employed in cardiovascular magnetic resonance (CMR) imaging are spin echo imaging, gradient echo imaging, late gadolinium enhancement (LGE) sequences, and flow velocity encoding. (See 'Techniques' above.)
●CMR uses no ionizing radiation. However, CMR is potentially problematic in patients with certain types of metallic implants. Thorough screening of all patients prior to CMRI is mandatory. (See 'Safety' above and "Diagnostic imaging in pregnant and lactating patients" and "Patient evaluation for metallic or electrical implants, devices, or foreign bodies before magnetic resonance imaging", section on 'Assessing implants, devices, or foreign bodies for MRI'.)
●Clinical applications of CMR include imaging of aortic, pericardial, myocardium and myocardial infarction, valvular, peripheral artery, congenital heart disease, ischemic heart disease, and intraventricular thrombus. (See 'Clinical applications' above.)
●CMR enables assessment of cardiac structure and function and vascular structures largely without need for an exogenous contrast agent. Gadolinium-based contrast agent administration is indicated for angiography or assessment of contrast enhancement properties of scar or infarction but should be avoided in patients with severe kidney disease. (See 'Kidney disease' above and "Nephrogenic systemic fibrosis/nephrogenic fibrosing dermopathy in advanced kidney disease".)
●Late gadolinium enhancement (LGE) is a technique that enables identification of myocardial fibrosis, including myocardial infarction and focal myocardial fibrosis, in patients with hypertrophic and other cardiomyopathies. Characteristic LGE patterns are helpful for diagnosis of amyloid cardiomyopathy, myocarditis, and cardiac sarcoid. Pericardial LGE is a sign of inflammation. (See 'Late gadolinium enhancement' above and 'Pericardial disease' above and "Cardiac amyloidosis: Epidemiology, clinical manifestations, and diagnosis", section on 'Cardiovascular magnetic resonance'.)
●CMR enables identification and quantification of myocardial iron overload, which may be helpful for diagnosis, prognosis, and treatment. (See 'Iron overload' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
