Cardiovascular magnetic resonance imaging

Cardiovascular magnetic resonance imaging

INVESTIGATIONS Cardiovascular magnetic resonance imaging Key points C Cardiovascular magnetic resonance imaging (CMR) has further advanced in the l...

1MB Sizes 0 Downloads 134 Views

INVESTIGATIONS

Cardiovascular magnetic resonance imaging

Key points C

Cardiovascular magnetic resonance imaging (CMR) has further advanced in the last 4 years, becoming faster and more patient-friendly

C

CMR is considered the gold standard technique for assessing cardiac anatomy, function and viability

C

First-pass perfusion imaging with CMR allows assessment of the presence and extent of perfusion abnormalities, which, in conjunction with infarct imaging, provides important information for planning revascularization strategies

C

The application of T1 and T2 mapping to quantitatively measure tissue characteristics has undergone further improvement and is now a robust technique applied in everyday clinical practice in many centres

C

In the clinical scenarios of potential iron overload, amyloidosis, AndersoneFabry disease and myocarditis, parametric mapping provides unique clinical information

C

Overall, CMR has a growing clinical role in ischaemic heart disease and various non-ischaemic cardiomyopathies, with a rapidly expanding pool of prognostic data

C

Coronary imaging is feasible with CMR, but the identification and grading of stenoses remains challenging

C

Although still at the research stage, molecular imaging may in future allow specific detection of unstable plaque

Theodoros Karamitsos Stefan Neubauer

Abstract Magnetic resonance imaging (MRI) uses the magnetic properties of the hydrogen nucleus, radiowaves and powerful magnets to provide highquality still and cine images of the cardiovascular system, with and without the use of exogenous contrast (gadolinium). Cardiovascular MRI (CMR) is the gold standard method for three-dimensional analysis of cardiothoracic anatomy, assessment of global and regional myocardial function and viability imaging (late gadolinium enhancement technique). Using first-pass perfusion imaging under vasodilator stress, CMR has high diagnostic accuracy for identifying myocardial ischaemia. Oedema imaging using T2-weighted techniques is useful for the identification of acute coronary syndromes and myocardial inflammation. Coronary MRI is feasible, and indicated particularly for visualizing anomalous coronaries. Its spatial and temporal resolution is inferior to computed tomography or conventional angiography, and the identification and grading of stenoses remains challenging. Molecular imaging may in future allow visualization of unstable plaque. Novel techniques such as T1 and T2 mapping offer a quantitative measure of tissue characteristics. CMR also provides important prognostic data for many cardiovascular diseases. CMR is now an essential component of an advanced cardiovascular imaging service, and it is anticipated that its role will continue to grow.

Keywords Cardiac anatomy; cardiac function; coronary angiography; diffuse fibrosis; magnetic resonance imaging (MRI); mapping; MRCP; oedema; perfusion; viability

Background Magnetic resonance imaging (MRI) is typically based on the magnetic properties of the hydrogen nucleus, although other nuclei can be used.1 In an MRI examination, the patient is placed in a powerful magnetic field that acts to align the body’s protons. Radiowaves in the form of a radiofrequency pulse transmitted into the patient cause the proton alignment to change, for example by 90 . When this radiofrequency pulse is turned off, the protons in the patient’s body return to their neutral position, emitting their own weak radiowave signals, which are detected by receiver coils and used to produce an image. The phase and amplitude of each returning radiowave signal can be determined using powerful computers and additional magnetic field gradients, and used to map the position of the excited protons. The resulting image reflects not only proton density, but also the highly complex manner in which protons resonate in their local environment. CMR requires advanced technology, including a high-field magnet (typically 1.5 Tesla (T), although recently 3.0 T systems have increasingly been used), fast-switching gradient coils and coils for transmission and signal reception. Compared with other imaging techniques, MRI has a unique ability to perform tissue characterization. Image contrast is influenced by proton

Introduction Cardiovascular magnetic resonance imaging (CMR) has undergone major technical progress over the last decade. CMR scanning has become faster and more patient-friendly, and image quality has further improved. A study of cardiac anatomy, (left and right ventricular) function and fibrosis with a modern CMR scanner can be performed in <30 minutes by an experienced operator. These improvements have led to the widespread adoption of CMR in clinical practice.

Theodoros Karamitsos MD PhD is an Assistant Professor of Cardiology at the Aristotle University of Thessaloniki, Greece. Research areas include stress perfusion and oxygenation CMR imaging, and the use CMR in the characterization of non-ischaemic cardiomyopathies. Competing interests: none declared. Stefan Neubauer MD FRCP FACC FMedSci is a Professor of Cardiovascular Medicine and Honorary Cardiology Consultant at the John Radcliffe Hospital, Oxford, UK. Research areas include most aspects of CMR as well as cardiac metabolism. Competing interests: none declared.

MEDICINE --:-

1

Ó 2018 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Karamitsos T, Neubauer S, Cardiovascular magnetic resonance imaging, Medicine (2018), https://doi.org/ 10.1016/j.mpmed.2018.05.008

INVESTIGATIONS

density and T1 and T2 relaxation times, which vary substantially for different tissues (Appendix Table 1). Another way to modify image contrast is by modulating how radiofrequency pulses are played out (the MRI sequence; Appendix Table 2). During an MRI scan, subjects and operators are not exposed to ionizing radiation, and there are no known detrimental biological adverse effects of MRI if safety guidelines are followed. The scanner attracts ferromagnetic objects, turning them into projectiles that could lead to significant patient or operator injury and also damage the scanner. The presence of certain medical implants and devices (e.g. pacemakers, defibrillators, cochlear implants, cerebrovascular clips) is a contraindication for routine MRI scanning, but nearly all prosthetic cardiac valves, coronary and vascular stents, and orthopaedic implants are safe in a 3 T (or less) MRI environment. Claustrophobia is a problem in a small percentage of patients, and mild sedation usually helps to overcome this. Gadolinium-containing contrast agents have been linked with the development of a rare systemic disorder called nephrogenic systemic fibrosis. Patients at risk of developing this disease are those with acute or chronic severe renal insufficiency (glomerular filtration rate <30 ml/minute/1.73 m2), or acute renal dysfunction of any severity caused by hepato-renal syndrome or in the perioperative liver transplantation period. To date, there is no evidence that other patient groups are at risk. Many MRI centres use gadolinium agents that are tightly bound to a cyclic chelate, which have an incidence of nephrogenic systemic fibrosis of nearly zero. However, it is not known whether immediate haemodialysis protects against nephrogenic systemic fibrosis, so gadolinium-based contrast media should be avoided in high-risk patients unless the diagnostic information sought using contrast-enhanced CMR is essential and not available with non-contrast-enhanced CMR or other imaging modalities.

Figure 1 A follow-up surveillance scan in a 75-year-old man who had previously undergone repair of a type A aortic dissection (ascending aorta interpositional graft). There is a small residual dissection in the aortic root. The chronic dissection continues distal to the graft, around the aortic arch and into the abdominal descending aorta. Source: Myerson S and Neubauer S, University of Oxford Centre for Clinical Magnetic Resonance Research.

extent of hypertrophy and fibrosis.1 Myocardial and liver iron overload in thalassaemia can be assessed quantitatively by measuring myocardial T2*.1 CMR also helps to differentiate constrictive from restrictive cardiomyopathy; in constriction, thickened pericardium on anatomical images and abnormal motion of the septum resulting from increased interventricular dependence can be readily recognized using real-time imaging during inspiration. By contrast, some patients with restrictive cardiomyopathy (e.g. from amyloidosis or endomyocardial fibrosis) can have areas of fibrosis on late gadolinium enhancement (LGE) CMR. Myocardial function and mass CMR is the accepted gold standard for quantification of myocardial mass and function. Using steady-state free precession techniques, double-angulated short-axis cine views can be obtained during all phases of the cardiac cycle (cine-MRI). Planimetry of each slice and summation of slice volumes allow precise determination of systolic and diastolic left and right ventricular volumes, stroke volumes, ejection fraction and myocardial mass, with high reproducibility (Figure 2). This is particularly important in patients with deformed hearts that have lost symmetry (e.g. after myocardial infarction), in whom Mmode and two-dimensional echocardiographic measurements (which assume left ventricular symmetry) are invalid. Because of the higher reproducibility, use of CMR instead of echocardiography in longitudinal clinical research studies of cardiac volumes and mass allows a several-fold reduction in patient group sizes. Analysis of regional myocardial function is feasible both at rest and during pharmacological stress, typically using dobutamine. Dobutamine stress CMR produces better quality images and thus has a higher sensitivity and specificity for detecting coronary artery disease than dobutamine stress echocardiography, particularly in patients with difficult acoustic windows. New CMR techniques (e.g. tagging, tissue phase-mapping) allow

Applications of CMR Normal and pathological anatomy Historically, the first application of CMR was the threedimensional analysis of cardiothoracic anatomy. By providing excellent soft tissue contrast, cardiovascular anatomy can be assessed in virtually any imaging plane (coronal, transverse, sagittal), including individualized doubleeoblique planes. The latter are particularly valuable in complex congenital heart disease. CMR has a very high degree of sensitivity and specificity for detecting diseases of the thoracic aorta such as aneurysm, acute dissection and intramural haemorrhage. It also allows investigation of the consequences of dissection (e.g. thrombosis, aortic incompetence, pericardial effusion) (Figure 1). Thoracic masses found on chest radiography or echocardiography are also indications for CMR, which can reveal their anatomical relationship to the normal cardiac and thoracic structures. In some cases, CMR allows tissue characterization of such masses. CMR is currently the most accurate method for diagnosing arrhythmogenic right ventricular cardiomyopathy, characterized by regional wall motion abnormality, regional wall thinning and, in advanced cases, fibrofatty infiltration.1 In hypertrophic cardiomyopathy, CMR can diagnose and determine the distribution and

MEDICINE --:-

2

Ó 2018 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Karamitsos T, Neubauer S, Cardiovascular magnetic resonance imaging, Medicine (2018), https://doi.org/ 10.1016/j.mpmed.2018.05.008

INVESTIGATIONS

Figure 2 End-diastolic still images from multiple contiguous short-axis steady-state free precession cines that encompass the left and right ventricle from base to apex. Note the position of the short-axis slices marked on the still frame of the end-diastolic horizontal long-axis cine image. Source: Karamitsos T and Neubauer S, University of Oxford Centre for Clinical Magnetic Resonance Research.

analysis of regional tissue contractility, although the clinical relevance of these methods remains to be shown.

A large number of clinical trials have assessed the feasibility, safety and diagnostic accuracy of stress perfusion CMR. A recent meta-analysis (2902 patients) showed that, compared with quantitative coronary angiography as the gold standard, firstpass perfusion CMR under vasodilator stress could diagnose coronary artery disease (70% stenosis) with excellent sensitivity (86%) and very good specificity (77%).2 The Clinical Evaluation of Magnetic Resonance Imaging in Coronary Heart Disease (CE-MARC) study compared the diagnostic accuracy of stress perfusion CMR with single-photon emission computed tomography (SPECT); it showed that both techniques had similar specificity, but CMR was more sensitive than scintigraphy for detecting ischaemia.3 It should be noted that CMR perfusion techniques have higher spatial resolution than nuclear techniques (by at least an order of magnitude) and can be used to study the transmural aspect of myocardial perfusion. However, although this higher resolution can demonstrate very small

Myocardial perfusion Regional myocardial perfusion can be measured after an intravenous bolus injection of an MRI contrast agent (gadoliniumDTPA), using ‘first-pass’ imaging. Using sequential multislice fast gradient-echo CMR (Figure 3), passage of the contrast agent through the heart chambers and the myocardial tissue can be followed. Regional timeeintensity curves can be derived from a series of such images. Pharmacological vasodilatation (adenosine, dipyridamole) induces a 3- to 5-fold increase in blood flow in myocardial areas subtended by normal coronary arteries, whereas no (or only minimal) change is found in areas subtended by stenotic coronary arteries. Thus, contrast arrival in these areas is delayed, and they appear hypo-intense (dark) compared with adjacent normal myocardium (Figure 3).

Figure 3 Two representative examples of CMR perfusion scans during adenosine stress. (a) This patient had a significant stenosis of the left anterior descending coronary artery. Note the area of hypo-enhancement in the anteroseptum and anterior wall (arrowheads); this represents a perfusion defect. (b) This subject has normal coronary arteries. Note the homogeneous enhancement in all myocardial regions during the infusion of adenosine. Source: Reprinted from Am Heart J; 162 (1), Karamitsos TD, Dall’Armellina E, Choudhury RP, et al., Ischemic heart disease: Comprehensive evaluation by cardiovascular magnetic resonance, 16e30, 2011, with permission from Elsevier.

MEDICINE --:-

3

Ó 2018 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Karamitsos T, Neubauer S, Cardiovascular magnetic resonance imaging, Medicine (2018), https://doi.org/ 10.1016/j.mpmed.2018.05.008

INVESTIGATIONS

Figure 4 A 65-year-old patient with previous inferior myocardial infarction referred for viability imaging. Panels (a) (end-diastolic still frame from short-axis cine) and (b) (end-systolic frame) show areas of akinesia in the inferior septum and left ventricular wall (white arrows) that are thinned. Note also the area of akinesia in the inferior right ventricular wall (black arrows). Panel (c) is the corresponding late gadolinium enhancement image, which shows full-thickness infarction involving the left ventricular inferior wall and inferior septum (white arrows), and the right ventricular inferior wall (black arrows). Source: Karamitsos T and Neubauer S, University of Oxford Centre for Clinical Magnetic Resonance Research.

perfusion defects not seen on nuclear imaging, the clinical implications remain to be established.

in cardiology. On inversion recovery T1-weighted sequences (Appendix Table 2) obtained 5e10 minutes after gadolinium administration, non-viable myocardium (scarred, irreversibly injured) shows high signal intensity, whereas normal and viable (stunned, hibernating) myocardium shows low signal intensity (Figure 4). The LGE technique has undergone extensive

Myocardial viability The determination of myocardial viability using gadolinium-based contrast agents (LGE technique) has revolutionized the use of CMR

Figure 5 Example of dilated (a, b) and hypertrophic (c, d) cardiomyopathy. Left panels, still end-diastolic frames of horizontal long-axis cines. Right panels, late gadolinium enhancement (LGE) images. For the patient with dilated cardiomyopathy, note the mid-wall striae of LGE in the septum (white arrows) and the presence of apical thrombus (white arrowhead). For the patient with hypertrophic cardiomyopathy, note the asymmetrically thickened septum, which shows diffuse, patchy LGE (white arrows). Source: Karamitsos T and Neubauer S, University of Oxford Centre for Clinical Magnetic Resonance Research.

MEDICINE --:-

4

Ó 2018 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Karamitsos T, Neubauer S, Cardiovascular magnetic resonance imaging, Medicine (2018), https://doi.org/ 10.1016/j.mpmed.2018.05.008

INVESTIGATIONS

histopathological validation. The superb spatial resolution of LGECMR allows the detection of even small subendocardial infarcts that might otherwise be missed by lower spatial resolution techniques such as SPECT. Several studies have demonstrated an inverse relationship between the transmural extent of myocardial infarction and recovery of segmental contractile function after revascularization. In practice, segments that show >50% scarring are considered non-viable, whereas segments with only subendocardial enhancement (<50%) have a high likelihood of functional recovery. Highly specific patterns of fibrosis and scarring have also been described for many non-ischaemic cardiomyopathies (Figure 5). Consequently, the LGE technique is a major part of nearly every scanning protocol and provides valuable diagnostic and pathophysiological insights.1 CMR can also assess myocardial viability using a low-dose dobutamine protocol in a way analogous to echocardiography, but in practice this is required only for patients with intermediate, i.e. around 50%, scarring. Magnetic resonance (MR) spectroscopy detecting 31phosphorus (high-energy phosphates e ATP and phosphocreatine) or 23 sodium (tissue sodium content) instead of hydrogen can provide intrinsic contrast (metabolic information) on viability without the need for a contrast agent. However, the spatial

resolution of non-hydrogen methods must be substantially improved before they can be considered clinically relevant. Myocardial oedema Various technical improvements have enabled the wide clinical use of T2-weighted CMR for the qualitative or semi-quantitative detection of myocardial oedema and inflammation, primarily in acute coronary syndromes and myocarditis.1 However, wellrecognized limitations of conventional T2-weighted techniques include the need for a ‘normal’ reference region of interest, either in remote myocardium or skeletal muscle; this can lead to falsenegative results when these reference areas are also affected in systemic processes. Novel quantitative T2 and T1 mapping techniques have been developed to overcome these limitations. Coronary arteries Using phase-contrast or time-of-flight methods, images can be obtained that show only those structures in which flow is present (MR angiography). For peripheral arteries, this technique has achieved excellent, clinically relevant resolution, but CMR of the coronary vessels remains a technical challenge because of their small size (up to 4 mm) and the continuous, complex movement.

Figure 6 CMR end-diastolic frame from cine (left panel), shortened modified look-locker inversion recovery (ShMOLLI) non-contrast T1 map (middle panel), and late gadolinium enhancement (LGE) images (right panel) in a healthy volunteer, aortic stenosis patient and cardiac amyloid patient. Note the markedly elevated myocardial T1 time in the cardiac amyloid patient (1170 ms, into the red range of the colour scale) compared with the normal control (955 ms) and the patient with aortic stenosis and left ventricular hypertrophy (998 ms). ED, end-diastolic. Source: Reprinted from Karamitsos TD, Piechnik SK, Banypersad SM, et al., Non-contrast T1 mapping for the diagnosis of cardiac amyloidosis. JACC Cardiovasc Imaging, 2013;6 (4):488e97, with permission from Elsevier.

MEDICINE --:-

5

Ó 2018 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Karamitsos T, Neubauer S, Cardiovascular magnetic resonance imaging, Medicine (2018), https://doi.org/ 10.1016/j.mpmed.2018.05.008

INVESTIGATIONS

Appendix

Fast, flow-sensitive gradient-echo sequences allow imaging of proximal coronary arteries using breath-hold or navigator techniques, with a maximum in-plane resolution of about 700 mm.4 MR coronary angiography can be used for the diagnosis of anomalous coronary arteries.4 However, the sensitivity for coronary stenosis is only 60e90% because of low spatial resolution. Further developments (parallel acquisition, gradient performance, intravascular contrast agents, higher field magnets) might in the future allow the development of high-resolution MR coronary angiography with CT-like quality. In the long term, the most important aspect of coronary MRI may be that the method allows the study of the coronary artery wall using T1-weighted, T2-weighted and spin density-weighted sequences. This may allow determination of the coronary plaque burden and, eventually, characterization of plaques as stable or unstable. The spatial resolution is currently insufficient for this application, but with technical progress it might become possible to identify patients at risk of coronary plaque rupture. Molecular contrast agents that specifically bind to components of unstable plaque may in future provide a solution for this, even if spatial resolution remains limited.

Relaxation times in MRI In MRI, two independent relaxation times are described with respect to the direction of the main magnetic field e longitudinal relaxation and transverse relaxation. Conventionally, these are termed the ‘T1’ and ‘T2’ relaxation times. Long T1 times reflect slower relaxation parallel to the main magnetic field; long T2 times reflect slower relaxation in the transverse plane. T1 values are typically several times greater than T2 values. T1 and T2 times vary considerably between different tissues, and these differences are the basis of much of the remarkable contrast resolution of MRI. Tissues with high water content have particularly long T1 and T2 times. T1 and T2 weighting C In T1-weighted images, areas with a long T1 time give a low signal. Water-rich areas therefore appear dark. C In T2-weighted images, areas with a long T2 time give a high signal. Water-rich areas therefore appear bright. T2* is a time constant describing the exponential decay of signal resulting from spinespin interactions, magnetic field inhomogeneities and susceptibility effects. T2* is measured by acquiring several T2-weighted images with different echo times. Shorter relaxation times (e.g. from iron loading) cause a more rapid decrease in myocardial signal intensity with increasing echo time, and this rate of decline can be plotted.

T1 and T2 mapping techniques T1 and T2 mapping refers to parametric maps generated from a series of images acquired with different T1 or T2 weighting so that each pixel can be assigned a T1 or T2 value, respectively.5 These maps can be displayed using colour scales to enable quantitative visual interpretation.5 Each tissue type exhibits a characteristic range of normal T1 and T2 relaxation times at a particular field strength, deviation from which can be indicative of disease. Myocardial T1 mapping methods are used for native (i.e. without the use of gadolinium-based contrast agents) and also post-contrast T1 measurements. In combination with haematocrit, these enable the quantification of extracellular volume (ECV fraction).5 Myocardial ECV can act as a surrogate marker of fibrosis when other pathologies that increase the extracellular space, such as myocardial oedema/inflammation, infiltration and ischaemia, have been excluded.5 Elevated T1 times and ECV in the myocardium have been reported in a number of commonly encountered cardiac conditions, including myocardial infarction, myocarditis, hypertrophic and dilated cardiomyopathy, cardiac amyloidosis (Figure 6), cardiac involvement in systemic diseases, and diffuse fibrosis in patients with aortic stenosis. Native myocardial T1 values can be lowered by watereprotein interactions, fat or iron content and thus can also serve as a diagnostic tool in characterizing AndersoneFabry disease, fat in cardiac masses and myocardial siderosis. T2 mapping can detect oedematous myocardial territories in a variety of cardiac pathologies, including acute myocardial infarction, myocarditis, Takotsubo cardiomyopathy and heart transplant rejection.

Table 1

MRI sequences An MRI sequence comprises a series of radiofrequency pulses that provide the MR signal. These are interleaved with a series of field gradient pulses, which provide the spatial encoding of the signal and hence the image. Spin-echo sequences have traditionally been the ‘workhorse’ of routine MRI. A 90 pulse is followed by a 180 pulse, and the delay between the two is reflected in the echo time (TE). The process is repeated after a repetition time (TR). C A spin-echo sequence with a short TR and a short TE produces an image in which long-T1 areas give a low signal (i.e. are black). C Use of long TR and long TE values produces an image in which long-T2 areas give a high signal (i.e. are white). C Long TR and short TE values produce a proton density-weighted image. As a ‘rule of thumb’, T1-weighted and proton density-weighted images tend to be similar to CT images and are particularly helpful for anatomical orientation. T2-weighted images can be more susceptible to artefacts, but are particularly sensitive to pathological lesions. Many diseased areas appear bright on T2-weighted images, partly as a result of their high water content. One limitation of spin-echo sequences is that they can be relatively slow. Fast spin-echo sequences are similar to conventional spin-echo sequences, but the data are collected faster. Image quality is greater for the same acquisition time, but artefact problems may be worse.

CMR and prognosis Data on the prognostic role of CMR in both ischaemic and nonischaemic cardiomyopathies are rapidly becoming available. The completion of ongoing multicentre trials is expected to provide more outcome and cost-effectiveness data that will further strengthen the clinical role of CMR. A MEDICINE --:-

6

Ó 2018 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Karamitsos T, Neubauer S, Cardiovascular magnetic resonance imaging, Medicine (2018), https://doi.org/ 10.1016/j.mpmed.2018.05.008

INVESTIGATIONS

coronary artery disease: a systematic review and meta-analysis. Int J Cardiol 2018; 252: 229e33. 3 Greenwood JP, Maredia N, Younger JF, et al. Cardiovascular magnetic resonance and single-photon emission computed tomography for diagnosis of coronary heart disease (CE-Marc): a prospective trial. Lancet 2012; 379: 453e60. 4 Bluemke DA, Achenbach S, Budoff M, et al. Noninvasive coronary artery imaging: magnetic resonance angiography and multidetector computed tomography angiography: a scientific statement from the American heart Association Committee on cardiovascular imaging and intervention of the Council on cardiovascular Radiology and intervention, and the Councils on clinical Cardiology and cardiovascular disease in the Young. Circulation 2008; 118: 586e606. 5 Messroghli DR, Moon JC, Ferreira VM, et al. Clinical recommendations for cardiovascular magnetic resonance mapping of T1, T2, T2* and extracellular volume: a consensus statement by the Society for Cardiovascular Magnetic Resonance (SCMR) endorsed by the European Association for Cardiovascular Imaging (EACVI). J Cardiovasc Magn Reson 2017; 19: 75.

Gradient-echo sequences: in gradient-echo sequences, the 180 pulse is replaced by a reversal of magnetic field gradients. This technique is generally much faster than conventional spin-echo, but can be more prone to artefacts. T1 weighting and T2* weighting can be undertaken. A refinement of this technique, now widely used to image cardiac function, is termed ‘steady-state free precession’ (SSPF); this provides the highest contrast between chamber blood (white) and myocardium (dark) of all available MR sequences. Echoplanar imaging: ultrafast techniques, particularly echoplanar imaging, are increasingly being used. They offer short imaging times (e.g. 30e40 ms per slice), but require advanced hardware. Inversion recovery sequences: in an inversion recovery sequence, a 180 pulse is followed by a 90 pulse after an interval TI. An important variant (short T1 inversion recovery (STIR) sequence) uses short TI values to suppress the signal from fat and highlight the signal from many diseased tissues. Another important example is the late enhancement sequence used to image myocardial viability. Navigator sequences: additional information can be acquired during the image sequence to enable correction for patient movements that would otherwise degrade the image. Use of this group of methods is now moving from the laboratory to newer clinical machines. Table 2

Acknowledgements The authors acknowledge support from the National Institute for Health Research Oxford Biomedical Research Centre Programme. Professor Stefan Neubauer also acknowledges support from the Oxford British Heart Foundation Centre of Research Excellence.

KEY REFERENCES 1 Karamitsos TD, Francis JM, Myerson S, et al. The role of cardiovascular magnetic resonance imaging in heart failure. J Am Coll Cardiol 2009; 54: 1407e24. 2 Kiaos A, Tziatzios I, Hadjimiltiades S, et al. Diagnostic performance of stress perfusion cardiac magnetic resonance for the detection of

TEST YOURSELF To test your knowledge based on the article you have just read, please complete the questions below. The answers can be found at the end of the issue or online here.

Question 1

What is the next most appropriate management plan? A Consider revascularization B Perform dobutamine stress CMR C Perform T1 mapping D Perform T2 mapping E Continue medical treatment

A 57-year-old man presented with tiredness and some shortness of breath on exertion. Three months previously, he had sustained an acute myocardial infection and was found to have three-vessel coronary artery disease. Cardiovascular magnetic resonance (CMR) imaging was performed for viability assessment.

Question 2 Investigations  Chest X-ray showed a normal-sized heart  ECG showed evidence of a previous anterior myocardial infarct  CMR showed a 25% wall thickness infarction in the anterior and septal wall of the left ventricle

MEDICINE --:-

A 34-year-old woman presented with acute severe central chest pain and troponin rise. Clinical examination was normal. Investigations  12-lead ECG was normal  Coronary artery angiography was normal

7

Ó 2018 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Karamitsos T, Neubauer S, Cardiovascular magnetic resonance imaging, Medicine (2018), https://doi.org/ 10.1016/j.mpmed.2018.05.008

INVESTIGATIONS

Investigations  ECG showed non-specific T-wave changes  Chest X-ray showed a normal-sized heart  Cardiovascular magnetic resonance imaging showed low native T1 images in the myocardium

Which MR sequence is most appropriate to detect myocardial oedema/inflammation in this situation? A. Cine steady-state free precession B. Inversion recovery T1-weighted C. T2-weighted spin-echo with short T1 inversion recovery (STIR) D. T1-weighted turbo spin-echo (TSE) E. T2* mapping

What is the most likely pathological change in the myocardium? A Deposition of glycosphingolipid B Amyloid accumulation C Chronic inflammation and fibrosis D Oedema E Ischaemia

Question 3 A 28-year-old man presented with tiredness, breathlessness on exertion and non-specific chest pain. On clinical examination, his heart rate was 88 beats/minute, blood pressure 148/94 mmHg, and jugular venous pressure not raised. There was a 2/6 short systolic murmur in the mitral area.

MEDICINE --:-

8

Ó 2018 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Karamitsos T, Neubauer S, Cardiovascular magnetic resonance imaging, Medicine (2018), https://doi.org/ 10.1016/j.mpmed.2018.05.008