CARDIAC IMAGING 

CARDIAC MR:  ONE-STOP SHOP

K.M.Das*, MD
Hamad Medical Corporation, Doha, Qatar

Introduction
Basics of Magnetic Resonance
  Motion Control:  ECG gating
  Respiratory gating
Clinical applications of basic and advanced MR techniques
  Ventricular function
  Ischemic heart disease 
  Myocardial viability
  Coronary Angiography
  Acute coronary syndrome
  Valvular heart disease
Multidetector CT (MDCT) vs CMR
Basics of Magnetic Resonance
Contraindications
Summary
References

Introduction 

    Cardiovascular disease is one of the leading causes of death in the United States. Every year $100 billion is spent on heart disease, much of which is used to reimburse diagnostic imaging studies1.  Selecting the most appropriate imaging modality, avoiding duplicate studies, and showing improved patient outcome based on cost-effective diagnostic and therapeutic algorithms are being demanded of radiologists. Echocardiography, nuclear cardiology and catheterization are gold standard techniques used all over the World. Over the last decade, significant work has been done to develop a test not only to replace these tests but also to have an additional advantage of “one-stop shop” evaluation. Cardiac magnetic resonance (CMR) has achieved a near gold standard reputation in the diagnosis of cardiac disease and it is one of the strongest contenders among competing modalities. 

1. Basics of Magnetic resonance

    Magnetic resonance imaging (MRI) is generally based on the detection of water and fat protons (1H) in the body. The human body has a high concentration of 1H, hence even weak magnetic resonance signals can be used to create a distribution map or image of this nuclide.

    MRI depends on the detection of the intrinsic angular momentum or spin of protons, which is a basic property of matter. 1H has one proton and no neutrons, giving it a net spin detectable in MRI.  The frequency of the spin associated with a nuclide is related to magnitude of the magnetic field (B0). A complex mathematical process called Fourier analysis produces the image.  It is beyond the scope of this article to describe more details of the examination process of the magnet and the image production and interested readers may refer to article published earlier2.

1.1. Motion control: ECG gating

    Unlike other parts of the body, the heart is an organ in continual motion and the motion is a significant challenge in cardiac magnetic resonance (CMR) imaging. The cylindrical motion of the heart, and of the diaphragm on which it sits, adds new constraints to CMR. Motion blurs the images obtained. To avoid blurring of images, acquisition of the image data must occur during intervals that are relatively short when compared with the time scale of cardiac and respiratory motion. 

    Significant progress in rapid CMR techniques has allowed acquisition of images with sufficient spatial resolutions during a fraction of a single cardiac cycle. However, in many cases more stringent requirement for spatial resolution and /or signal-to-noise ratio (SNR) requires longer acquisition. These constraints have led to the routine use of ECG triggering in most CMR scans. 

    In general, the imaging time is defined by two parameters: the acquisition window within each cardiac cycle and the total number of cardiac cycle. In a patient with a heart rate of 60 beats per minute, each cycle accounts for 1000 ms (millisecond). To avoid blurring from the cardiac motion, a window of less than 50ms or less is selected in between the RR interval. Bulk cardiac motion is most rapid during ventricular systole, but also following atrial systole, with relatively little motion during isovolumic ventricular relaxation and mid-diastole. To identify that specific time in the cardiac cycle, it is necessary to get an accurate tracing of the RR interval. Invariably, the magnetic gradients and magneto-hydrodynamic effect of systolic aortic flow causes surface potentials on the ECG that distorts the ST-T segment to produce a tall T wave, which may be misinterpreted as R wave. Recently, Fisher and colleagues have described a magnetic resonance (MR) compatible technique using vector cardiography and this advance has been proven to be very useful3. The system examines the 3-D orientation of the ECG signal and uses the calculated vector of the QRS complex as a filter mechanism to ignore electrical signals that are of a similar timing in the cardiac cycle or a similar magnitude, but of a different vector.

 

1.2. Respiratory gating

    The cardiac application of CMR has further improved with the maturity of the new techniques to control motion and flow artifacts. Respiratory motion responsible for the significant blurring of the CMR image has been controlled with the development of fast breath-hold imaging technique and 3-D navigator free-breathing technique. This has moved CMR out of traditional cardiac gated spin-echo techniques into the newer segmented breath-hold or navigator techniques.

2. Clinical applications of basic and advanced MR techniques

    With further development of the MR hardware and gradient amplifier enhancements with shorter rise time (slew rate ~ 200 mtesla/m/msec), CMR can perform superior cardiac studies. In addition, development of the high-sensitive RF coils and digital processing algorithms has further enhanced the speed image acquisition and reconstruction during cardiac studies. New parallel data acquisition techniques like SMASH (simultaneous acquisition of spatial harmonics) or SENSE (sensitivity encoding) and use of phased-array multi-channel coil technology has further improved spatial and temporal resolution4,5. Recently, real-time cine MRI with frame rates up to 40 images per second has been in use using echo-planer imaging sequences with SMASH imaging5. With all these advances, it is possible in a clinical setting to evaluate cardiac morphology, function, dynamic perfusion; myocardial viability information; and even coronary anatomy and flow in one sitting. 

    CMR provides images with high contrast between the blood pool and the vascular wall without using contrast media. The two basic techniques used in CMR are Spin-echo and gradient-echo sequences. In Spin-echo, blood appears black because of slice selective 90° - 180° pulse pair and it is known as “black blood” technique (Fig.1). This technique is very useful in visualizing cardiac morphology in congenital heart disease, right ventricular dysplasia and pericardial abnormalities (Fig.1). 

Figure 1:  The pericardium is thickened (>4mm) (Arrow).

In gradient-echo, the appearance of the blood becomes brighter because of repetition of alpha pulses leading to saturation of the flowing blood as well as the stationary tissue (Fig 2). This sequence is widely used for the assessment of the coronary arteries, ventricular function, myocardial perfusion and qualitative valvular function. Its superior spatial and contrast resolution provides excellent anatomical knowledge of the heart and pericardium (Fig.3) in different planes as well as its surroundings. This technique provides high-resolution three-dimensional views of the cardiac anatomy with no interference from adjacent bone or air, yielding images superior to those of echocardiography. CMR is better than other techniques at revealing heart muscle contractility and function, as well as areas of ischemia or infarct.

Figure 2:  Black blood technique.  Blood inside the left ventricle and right ventricle appears black.

Figure 3:  White blood technique.  Blood inside the Left and right ventricle appears white.

2.1. Ventricular function

    The evaluation of ventricular function is one of the important assessments in patients with heart disease. CMR has made it possible to acquire short-axis cine MRI of the beating heart in real-time with high spatial and temporal resolution. Evaluation of cardiac function is possible with a series of shot-axis cine images acquired from the apex to the base of the heart in a fraction of a second and accurately measure left ventricular mass, end diastolic volume (EDV), end systolic volume (ESV), and ejection fraction (EF) (Fig.4). 3-D echocardiography has been available as an investigational technique for well over a decade. Recent investigators have found an excellent correlation between real-time 3-D echocardiography measurements of LVEF and those obtained by CMR6.

Figure 4:  White blood technique showing the short axis cross-section of LV from apex to the base

2.2. Ischemic heart disease

    Cardiac magnetic resonance is one of the reliable tools used for the diagnosis of ischemic heart disease. Apart from evaluation of the coronary arteries, there are two main techniques used to identify cardiac ischemia. One is the use of dobutamine to identify the areas of wall motion abnormality with cine gradient-echo imaging; the other technique employs gadolinium-enhanced first pass cardiac studies (Fig.5) to detect difference in perfusion following administration of persantine or adenosine.

Figure 5:  Stress CMR with adenosine.  
A. Post gadolinium perfusion reveals no perfusion defect in the subendocardium.

Fig 5B:  Post adenosine infusion reveals a perfusion defect (arrow) in the subendocardium.

Fig 5C:  Perfusion defect enhances in the delayed scan.  
(Courtesy Time Jones, LSI, Leeds, England)

    Considerable amount of work has been published supporting the efficacy of the stress CMR in the detection of inducible ischemia. Both dobutamine and adenosine have been widely used with good correlation with quantitative angiography. Ingkanisorn et al, reported high sensitivity and specificity of adenosine perfusion CMR for the detection of CAD in troponin-negative patients who had a recent emergency room visit to evaluate chest pain7. In this study of 141 patients, the sensitivity of adenosine perfusion CMR was 100% and the specificity was 91% when delayed hyper-enhancement for myocardial viability was combined with adenosine perfusion as well as the presence of regional wall motion abnormalities at rest. This combined approach yielded a sensitivity of 100% and specificity of 89%. One of the reasons for the high sensitivity of CMR compared to positron emission tomography (PET) and single photon emission tomography (SPECT) is the high spatial resolution on the order of 2mm or better. Dill et al, in their study of 71 patients, compared the diagnostic value of stress CMR for the detection of coronary artery disease to that of SPECT. The sensitivity of CMR for the detection of coronary disease compared well with SPECT (both about 85%) although the specificity of CMR was greater (96% vs. 74%). Moreover, the CMR was able to differentiate transmural from the subendocardial perfusion defects due to its high spatial resolution, which could not be done with SPECT8. Judd et al, in a comparative evaluation of 91 patients of known or suspected coronary artery disease found that SPECT detected only 53% of the micro-infarcts detected by CMR9. 

2.3. Myocardial viability

    In patients with coronary artery disease, the distinction between reversible and irreversible myocardial injury is essential. Recent studies indicate that CMR after administration of contrast material can be used to distinguish between reversible and irreversible myocardial ischemic injury. One of the important advantages of contrast-enhanced MRI over other imaging methods that are used to assess myocardial viability is that it shows transmural extent of viable myocardium. The transmural extent of the hyperenhancement was significantly related to the likelihood of improvement in contractility after revascularization. Myocardium that is ischemic but viable shows contractility in the cine image and may or may not show hyperenhancement on the perfusion images, with no evidence of delayed hyper-enhancement. Results of recent studies support the concept of delayed hyperenhancement, which represents predominantly infracted nonviable myocardium. Kim et al, demonstrated improvement of contractility in 78% of segments with no-hyperenhancement, but in only 1 of 58 segments with hyperenhancement of more than 75% of the tissue10. 

2.4. Coronary Angiography

    Magnetic resonance angiography (MRA) has evolved to be one of the noninvasive techniques for the diagnosis of CAD (Fig.6). 

Figure 6:  The whole length of the right normal coronary artery (arrow).  
Courtesy Tim Jones, LSI, Leeds, England     

    Coronary magnetic resonance angiography has already been shown as a useful tool to demonstrate of the anomalous origins, and it is often superior to x-ray angiography in the delineation of the course11. With the development of navigator gating free-breathing 3-D motion correction coronary magnetic resonance angiography, several investigators have come up with promising results for the evaluation of proximal and mid coronary artery disease12,13. Kim et al, in a multicenter study of 109 subjects of CAD, had an overall accuracy of 72% in diagnosing coronary artery disease14. The sensitivity, specificity and accuracy for patients with disease of left main coronary or three vessel disease were 100%, 85% and 87% respectively. The negative predictive values for any coronary disease and for left main artery or three-vessel disease were 81% and 100% respectively. In general, MRA would detect 94% of all patients with any coronary or with left main coronary or three-vessel disease. Delineation of left circumflex disease was less reliable but isolated disease of this artery is less than 4% and absence of clinically significant disease in the remaining coronary makes the left circumflex disease unlikely.

 2.5. Acute coronary syndrome 

    The management of chest pain in the emergency department remains a challenge with current diagnostic strategies. Acute coronary syndrome accounts for 2 million hospitalization per year in the United States alone. Mortality is approximately 25%, including those patients who never reach a hospital. CMR can be used for comprehensive evaluation of patients with acute coronary syndrome in less than 1 hour. Kwong et al, in a prospective study of 161 patients with the possible or probable diagnosis of acute coronary syndrome, CMR performed within 12 hours was more sensitive than strict ECG criteria for ischemia (p<0.001), peak troponin-I (p<0.001) and TIMI risk score (p=0.004). CMR was more specific that an abnormal ECG (p<0.001)15.

    One of the concerns with CMR stress studies has been the ability to handle emergency situations while the patient is inside the magnet. The most crucial difference when compared with conventional exercise is the lack of diagnostic ECG, precluding the assessment of stress induced ST segment changes. In the modern commercial scanner, with the introduction of vectorcardiogram and in-house blood pressure monitoring with closed circuit TV and presence of an experienced physician, the problem has been significantly brought under control. Several publications on stress CMR covering more than 6000 patients came out with the consensus that the overall safety was similar to that of dobutamine stress echocardiography7,16.

2.5a. BOLD MRI

    Bauer et al published a new risk free and contrast free MRI method based on “blood oxygenation level dependent” (BOLD), to identify the susceptible myocardium supplied by a stenotic coronary artery. The regional deoxyhemoglobin level is mapped by MRI phase relaxation (or T2*) times in hearts. The T2* value in the underperfused myocardium was 31% lower than the normal perfused area and this difference increased to 41% under stress condition, driven by increases in T2* in normal myocardium after dipyridamole infusion17.

2.5b. MR spectroscopy (MRS)

    MR spectroscopy (P-31 MRS) is also highly sensitive in the detection of myocardial viability. High-energy phosphate metabolites and inorganic phosphate ratio has been found to be very useful in the detection of ischemic heart disease and to see the results of coronary angioplasty. Within seconds after reduction of oxygen supply, a decrease in phosphocreatinine (PCr) and increase in inorganic phosphate (Pi) occur, the earliest metabolic response in the myocardium to ischemia18.

2.6. Valvular heart disease

    CMR is rapidly gaining acceptance as an accurate, non-invasive method for optimal assessment of structural and functional parameters in patients with valvular heart disease (Fig.7). Recent reports have demonstrated that dynamic magnetic resonance imaging may serve as an attractive alternative or complement to echocardiography19,20. It also provides 3-D anatomic and functional data and it is potentially more accurate in assessment of ventricular function than is possible with echocardiography21. Velocity encoded cine CMR imaging allows accurate estimation of velocity profiles across a valve or any vascular structure, thereby providing quantitative measurements comparable to those provided by color flow Doppler echocardiography22,23.

Figures 7 A & B: Four chamber view showing both open (arrows) (A) and 
closed (arrows) (B) mitral and tricuspid valves.

3. Multidetector CT (MDCT) vs CMR

    Cardiac multidetector computed tomography (MDCT) is one of the strongest competing modality with significant promise. Currently, MDCT angiography offers potential to supplant contrast coronary angiography. Gerber et al compared MDCT with 3-D CMR for the detection of coronary artery stenosis in 45 patients who had coronary angiography24. The sensitivity of CT was slightly better than that of CMR (100% vs 89%), while the specificity was similar in both (70%).

    Contrast enhanced MDCT compares well with that of gadolinium based CMR for the detection of ischemia with minor difference in the size of the perfusion defect and post contrast enhancement. Several articles have shown that the higher the calcium score of the coronaries, the greater the risk of cardiovascular events25. Although the results look promising, the issue of the radiation dose for CT evaluation for a whole range of morphology, LV function, coronaries, perfusion, delayed enhancement of ischemic heart disease, including occasional multiple pass studies remains to be worked out. Nevertheless, MDCT techniques are somewhat easier to perform and interpret than many of the recently launched CMR applications. For this reason, MDCT may eventually gain wider clinical acceptance for noninvasive cardiac imaging, particularly coronary angiography, than apparently more complex CMR application.

 4. Contraindications

There are number of circumstances in which CMR is better avoided because of reports of death or harm to the patients. Cerebral aneurysmal clips may get dislodged and cause fatal hemorrhage. Although, modern clips are non-ferromagnetic and safe, a prior written specification of the clip is always necessary to avoid such catastrophe. Electronic implants may malfunction or get damaged by CMR. This applies to cochlear implants, automatic cardioverter defibrillators, nerve stimulation units and number of other implants. Coronary stents are mostly non-ferromagnetic and safe under CMR. Similarly the recent valvular prostheses are safe except for pre-6000 series Starr-Edwards valves. Patients with pacemakers should not be scanned unless special circumstances arise, and then only in centers with special CMR expertise, monitoring, and cardiology backup.

Summary

    The concept of “one-stop shop” is now becoming a reality and has been reported by several groups who performed examination in patients with myocardial infarction in less than 1 hour26,27. Kramer et al performed comprehensive CMR imaging assessment of LV structure and function, infarct artery patency, and regional myocardial contrast material uptake in 27 patients of acute myocardial infarction27. Lauerma et al, even combined dobutamine stress cine, first pass and late contrast enhancement in the same group of patients with multivessel disease28. With recent advances in the speed of MR technology, it is now feasible to integrate the examination of all cardiac problems at one time, which previously required as many as three different clinical cardiac imaging methods. 

    In conclusion, CMR is unique in the variety of its application in imaging the cardiovascular system: a true one-stop shop assessment of myocardial structure, function and perfusion and coronary anatomy and flow. With increasing accessibility of MR imagers as well as the growing availability of new hardware, more advanced techniques, and faster imaging sequences, CMR imaging will become a routine procedure for the investigation and follow-up of patients with cardiovascular disease.¨
The author thanks Dr. Jassim Al-Suwaidi for reviewing the manuscript.

References

1.  American Medical Association. Current procedural terminology (CPT) 1999. Chicago:      American Medical Association, 1999.

2.  Lauterbur P. Image formation by reduced local interactions: Examples employing nuclear      magnetic resonance. Nature. 1973; 242:190-201.

3.  Fisher SE, Wickline SA, Lorenz CH. Novel real-time R-wave detection algorithm based on the      vectorcardiogram for accurate gate magnetic resonance acquisitions. Magn Reson Med.1999;
     42,361.

4.  Weiger M, Pruessmann KP, Boesiger P. Cardiac real-time imaging using SENSE. Magn      Reson Med. 2000; 43:177-184.

5.  Sodicson DK, Grisworld MA, Jakob PM. SMASH imaging. MRI Clin N Am. 1999; 7(2): 
     237- 254.

6.  Sugeng L, MacEneaney P, Weinert L. Validation of real-time three-dimentional      echocardiographic measurements of left ventricular ejection fraction with magnetic resonance      imaging. Circulation 2003; 108 (Supl IV): IV-338. Abstract 1583.

7.  Ingkanisorn WP, Kowng RY, Rhoads KL. Diagnostic and prognostic value of adenosine stress      MRI in the assessment of chest pain in troponin-negative patients. Circulation. 2003; 108      (Suppl IV): IV-402.

8.  Dill T, Ekinci O, Hansel J. The diagnostic value of asenosine stress MRI in routine clinical      setting compared to SPECT and coronary angiography. Circulation. 2003; 108 (Suppl IV):      IV-402. Abstract 1859.

9.  Wanger A, Mahrholdt H, Holley TA, Elliot MD, Regenfus M, Parker M, Klocke FJ, Bonow RO,      Kim RJ, Judd RM. Contrast-enhanced MRI and routine single photon emission computed      tomography (SPECT) perfusion imaging for detection of subendocardial myocardial infarcts:       an imaging study. Lancet. 2003; 361: 374-379.

10. Kim RJ, Wu E, Rafel A, Chen EL. The use of contrast-enhanced magnetic resonance imaging      to identify reversible myocardial dysfunction. MEJM. 2000; 343:144-153.

11. McConnell MV, Ganz P, Selwyn AP, W Li. Identification of anomalous coronary arteries and      their anatomic course by magnetic resonance angiography. Circulation. 1995; 92: 3158-3162.

12. Stuber M, Botnar RM, Danias PG. Double-oblique free breathing high resolution 3-D coronary      magnetic resonance angiography. J Am Coll Cardiol. 1999; 34: 524-531.

13. Stuber M, Botnar RM, Danias PG. Submilimeter three-dimensional coronary MR angiography      with real-time navigator correction: comparison of navigator locations. Radiology. 1999;      212:579-587.

14. Kim WY, Danias PG, Stuber M, Flamm SD, Plein Sven et al, Coronary magnetic resonance      angiography for the detection of coronary stenosis. MEJM. 2001; 345 (26): 1863-1869.

15. Kwong RY, Schussheim AE, Rekhraj S. Detecting acute coronary syndrome in emergency      department with cardiac magnetic resonance imaging. Circulation. 2003; 107: 531-537.

16. Von der Recke G, Troatz C, Engels T, Levenson B. The clinical feasibility of cardiac stress      magnetic resonance imaging. Circulation. 2003; 108 (Supl IV): IV-453. Abstract 2083.

17. Christian M. Wacker, Andreas W. Hartlep, Stefan Pfleger, Lothar R. Schad, Georg Ertl,      Wolfgang R. Bauer. Susceptibility-sensitive magnetic resonance imaging detects human      myocardium supplied by a stenotic coronary artery without a contrast agent. J Am Coll      Cardiol. 2003; 41(5): 834-840.

18. Bottomly PA, Herfkens RJ, Smith LS, Bashore TM. Altered phosphate metabolism in      myocardial infarction: P-31 MR spectroscopy. Radiology. 1987; 165: 703-707.

19. Globits S, Higgins CB. Assessment of valvular heart disease by magnetic resonance       imaging. Am heart J. 1995; 129: 369-381.

20. Duerinckx AJ, Higgins CB. Valvular heart disease. Radiol clin North Am. 1994; 32: 613-630.

21. Higgins CB, Sakuma H. Heart disease: functional evaluation with MR imaging. Radiology.      1996; 199: 307-315.

22. Mostbeck GH, Caputo GR, Higgins CB. MR measurement of blood flow in cardiovascular      system. Am J Roent.1992; 159: 453-461.

23. Rebergen SA, van der Wall EE, Doornbob J. Magnetic resonance measurements of velocity      and flow: technique, validation and cardiovascular applications. Am Heart J. 1993; 126:
     1439-1456.

24. Gerber BL, Coche E, Pasquet A. Direct comparison of multislice coronary CT and free      breathing navigator 3D MRI for detection of coronary artery stenosis in patients. Circulation.
     2003; 108 (Suppl IV): IV-488. Abstract 2238.

25. Schmermund A, Dentkas AE, Rumberger JA. Independent and incremental value of coronary      artery calcium for predicting the extent of angiographic coronary artery disease: comparison
     with cardiac risk factors and radionuclide perfusion imaging. J Am Coll cardiol. 1999; 34:       777-786.

26. Kramer CM. Integrated approach to ischemic heart disease: the one stop shop. Cardiol Clin.      1998; 16: 267-276.

27. Kramer CM, Rogers WJ, Geskin G. Usefulness of magnetic resonance imaging early after      acute myocardial infarction. Am J Cardiol. 1997; 80: 690-695.

28. Lauerma K, Niemi P, Hanninen H. Multimodality MR imaging assessment of myocardial      viability: combination of first-pass and late contrast enhancement to wall motion dynamics and      comparison with FDG PET- initial experience. Radiology. 2000; 217: 729-736.

*Consultant radiologist Hamad Medical Corporation, Doha, Qatar.

Address for correspondence: Dr. K.M.Das, Department of Radiology, 
Hamad Medical Corporation, Po Box 3050, Doha, Qatar, Email: daskm@qatar.net.qa