Software analyzes MR first-pass perfusion data

November 1, 2006
Christopher Klassen, MD
Christopher Klassen, MD

,
Minh Nguyen, MD
Minh Nguyen, MD

,
Franz Von Ziegler, MD
Franz Von Ziegler, MD

,
Norbert Wilke, MD
Norbert Wilke, MD

Quantitative analysis of myocardial defects could move into routine clinical use with advances

Quantitative analysis of myocardial defects could move into routine clinical use with advances

At the University of Florida, we use multiple modalities, including echocardiography, nuclear stress perfusion, coronary CT angiography, and MR first-pass perfusion, for initial noninvasive detection of coronary artery disease. CTA routinely evaluates low to intermediate risk patients. The CT data are evaluated for extent and location of significant CAD through multiplanar reconstruction and 3D volume rendering. Once CAD is detected, further evaluation with additional noninvasive tests or coronary angiography is appropriate if significant stenoses have been noted on CT.

MR first-pass perfusion (MRFPP) imaging often follows coronary CTA to determine the physiologic effects and extent of decreased myocardial blood flow due to stenosis. Our approach to MR perfusion imaging uses software solutions developed from years of research in this area. Input data are the first-pass signal intensity curves from both the left ventricle blood pool and the myocardium. Output data are absolute myocardial blood flows and myocardial perfusion reserves.

A typical scenario involves a patient with a prior coronary CTA that either detected or failed to determine a stenosis. Often in the latter case, the patient has severe calcium. Any approach to cardiac imaging must take account of the global picture of the heart, using all the elements of a standard cardiac examination, including cine wall motion, wall thickness, and viability.

The presence of coronary stenosis is detected on MRFPP by determining diminished blood flow to the myocardium. Blood flow is imaged by using an injection of gadolinium contrast agents, which have yet to achieve FDA approval for cardiac use. The following synopsis represents an off-label use of Magnevist. Gadopentetate dimeglumine (Magnevist, Berlex) is injected via power injector (MedRad) at a dose of 0.1 mol/kg at a rate of 9 mL/sec followed by a normal saline flush. The scan is started immediately before injection to capture the first pass of contrast through the myocardium and continues another 50 to 70 seconds. Typically, the stress portion is acquired before the rest examination to maximize contrast enhancement and to eliminate the effect of early delayed hyperenhancement. The second contrast injection typically has persisting contrast in the blood pool and myocardium, depending on pathology, that may interfere with image interpretation and analysis.

Adenosine used for the stress portion of the examination is infused for four minutes at a dose of 140 mcg/kg/min. It causes dilation of the coronary smooth muscle cells through its effects on the A2a receptor. Adenosine in the normal heart increases blood flow to approximately four times that at rest. It has a half-life of less than four to 10 seconds, so its effects rapidly dissipate.

Wilson et al have shown that coronary stenosis of 50%, 70%, or 80% decreases blood flow by 33%, 55%, or 67%, respectively. Hence, MRI perfusion imaging must detect about a 50% decrement in blood flow to determine stenosis greater than 70%. Compensatory vasodilation distal to the stenosis may be able to maintain blood flow during resting conditions. But introducing a pharmacologic agent such as adenosine or dipyridamole that causes dilation of the distal vessels may exceed the capacity of this physiologic response.

The ratio of myocardial blood flow at stress and at rest is defined as the myocardial perfusion reserve. This ratio decreases with increasing stenosis severity. Using the vasodilator adenosine as the stress agent, the expected perfusion reserve is greater than 3.5 in normal volunteers. Significant stenosis ( > 75% luminal stenosis) decreases the myocardial perfusion reserve to less than 1.5.

SEQUENCES AND PROTOCOLS

Although a number of sequences are used for perfusion imaging, the two most common are saturation recovery fast low-angle single-shot (SR-FLASH) and saturation recovery echo-planar imaging (SR-EPI). Requirements of a perfusion sequence are fast acquisition, high signal to noise ratio, minimal artifacts, and high enough resolution to distinguish the subendocardium. Our group uses SR-FLASH sequences to image the first pass of contrast through the heart. Depending on heart rate, three to four slices are obtained, typically three short-axis images and one long-axis image.

A radiologist or cardiologist eyeballs the images, looking at the first pass of contrast through the myocardium. The qualitative approach is enhanced by using it systematically with specific criteria in mind. Our criteria are the following: the perfusion defect should not fluctuate in signal intensity during the first pass; the defect is located in the distribution of a coronary artery; and the defect is seen only during stress imaging or is present during rest, stress, and delayed contrast imaging.

Applying the converse of all these criteria aids in the detection of artifacts in the study. The converse would be signal intensity that fluctuates during the first pass, is present during rest but not stress, or is not present consistently in all studies.

We also apply quantitative analysis to the first-pass signal intensity curves. Using a software solution developed and validated for the quantitative analysis of myocardial perfusion studies, we derive absolute myocardial blood flow as well as the myocardial perfusion reserve. The first step in quantitative analysis is using a program to contour the myocardium. We use several programs, including Medis (Leidin, the Netherlands) and Pie Medical (Maastricht, the Netherlands). Contours are placed around the regions of interest by outlining the endocardial and epicardial surfaces. The short axis is then radially segmented by placing a reference point at the anterior intersection of the right and left ventricles. The standard segmentation of the myocardium is typically used to divide the myocardium into six radial sectors, although it may be divided into more sectors for research. The program then calculates the average signal intensity in each radial sector and outputs this in either numerical or graphic form.

The most significant limitation of quantitative methods is the need to accurately and consistently draw all the myocardial contours for all the time points in the first-pass perfusion sequence manually. Automated computer detection of myocardial contours continues to be an active area of research, and its implementation is widely anticipated in cardiac imaging.

The second step is to use a custom software program developed in C++ code. This technique has previously been derived and validated in animals using radioactive microspheres. Our program takes the signal intensity curves and calculates myocardial blood flow and perfusion reserve using a Fermi function deconvolution method (see figure). The files for the signal intensity curves are specified for rest and stress, and the program calculates a least squares fit to the data on the order of milliseconds. Having fit the data, the program derives the absolute myocardial blood flows and perfusion reserves. This information is then displayed graphically and also in a numerical table.

QUANTITATIVE ANALYSIS

Quantitative analysis is useful for a number of different clinical senarios. In patients with CAD, this analysis can help differentiate which coronary territories are affected. Cutoff values for myocardial perfusion reserve can be developed and applied that can assist in determining the severity of CAD stenosis. This analysis can be used for more accurate follow-up in comparing repeat studies in the same patient and can also be applied to microvascular obstruction. Wu has demonstrated that the presence of microvascular obstruction following myocardial infarction has prognostic information. Quantitative analysis can be used to assess arteriopathy in heart transplant patients.

In addition to fully quantitative measures, other semiquantitative indices can be derived from the signal intensity curves. These include the slope of signal intensity rise, peak of the signal intensity curve, and time to peak of the signal intensity curve. A decrease in slope of the signal intensity curves could indicate a decrease in myocardial blood flow. A myocardial perfusion reserve index, but not absolute blood flows, can be calculated using these techniques.

If a perfusion defect is detected, the next question is which artery is supplying that territory. We divide the short-axis perfusion myocardium into six to eight radial sectors and assign each sector to an artery on an anatomic distribution. In general, the anterior and anterior septal sectors are assigned to the left anterior descending artery, the lateral sectors to the left circumflex, and the inferior and inferior septal sectors to the right coronary artery. If the patient has had prior CTA or interventional catheterization reports and images, these are also reviewed.

Cardiac imaging has progressed immensely in recent years, in large part from postprocessing software and hardware solutions. MR perfusion imaging with sophisticated automated edge detection could move quantitative analysis into a routine clinical application in the future.

Dr. Klassen is a resident, Dr. Nguyen is a resident, Dr. von Ziegler is a research fellow, Mr. Siuciak is an imaging systems specialist, and Dr. Wilke is director of cardiovascular MR and CT, all at the University of Florida at Jacksonville.

References

1. Pennell DJ, Sechtem UP, Higgins CB, et al. Clinical indications for cardiovascular magnetic resonance (CMR): Consensus panel report. J Cardiovasc Magn Reson 2004;6:727-765.

2. Wilson RF, Marcus ML, White CW. Prediction of the physiologic significance of coronary arterial lesions by quantitative lesion geometry in patients with limited coronary artery disease. Circulation 1987;75:723-732.

3. Klocke FJ, Simonetti OP, Judd RM, et al. Limits of detection of regional differences in vasodilated flow in viable myocardium by first-pass magnetic resonance perfusion imaging. Circulation 2001;104:2412-2416.

4. Uren NG, Melin JA, De Bruyne B, et al. Relation between myocardial blood flow and the severity of coronary-artery stenosis. NEJM 1994;330:1782-1788.

5. Wagner A, Mahrholdt H, Holly TA, et al. 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.

6. McCrohon JA, Lyne JC, Rahman SL, et al. Adjunctive role of cardiovascular magnetic resonance in the assessment of patients with inferior attenuation on myocardial perfusion SPECT. J Cardiovasc Magn Reson 2005;7:377-382.

7. Budoff MJ, Cohen MC, Garcia MJ, et al. ACCF/AHA clinical competence statement on cardiac imaging with computed tomography and magnetic resonance: a report of the American College of Cardiology Foundation/American Heart Association/American College of Physicians Task Force on Clinical Competence and Training. J Am Coll Cardiol 2005;46:383-402.

8. Weinreb JC, Larson PA, Woodard PK, et al. American college of radiology clinical statement on noninvasive cardiac imaging. Radiology 2005;235:723-727.

9. Nagel E, Lorenz C, Baer F, et al. Stress cardiovascular magnetic resonance: consensus panel report. J Cardiovasc Magn Reson 2001;3:267-281.

10. Wolff SD, Schwitter J, Coulden R, et al. Myocardial first-pass perfusion magnetic resonance imaging: a multicenter dose-ranging study. Circulation 2004;110:732-737.

11. Kraitchman DL, Chin BB, Heldman AW, et al. MRI detection of myocardial perfusion defects due to coronary artery stenosis with MS-325. J Magn Reson Imaging 2002;15:149-158.

12. Chan SY, Brunken RC, Czernin J, et al. Comparison of maximal myocardial blood flow during adenosine infusion with that of intravenous dipyridamole in normal men. J Am Coll Cardiol 1992;20:979-985.

13. Cerqueira MD. The future of pharmacologic stress: selective A2A adenosine receptor agonists. Am J Cardiol 2004;94:33D-40D, discussion 40D-2D.

14. Zhang H, Shea SM, Park V, et al. Accurate myocardial T1 measurements: toward quantification of myocardial blood flow with arterial spin labeling. Magn Reson Med 2005;53:1135-1142.

15. Al-Saadi N, Nagel E, Gross M, et al. Improvement of myocardial perfusion reserve early after coronary intervention: assessment with cardiac magnetic resonance imaging. J Am Coll Cardiol 2000;36:1557-1564.

16. Jerosch-Herold M, Wilke N, Stillman AE. Magnetic resonance quantification of the myocardial perfusion reserve with a Fermi function model for constrained deconvolution. Med Phys 1998;25:73-84.

17. Jerosch-Herold M, Swingen C, Seethamraju RT. Myocardial blood flow quantification with MRI by model-independent deconvolution. Med Phys 2002;29:886-897.

18. Wu KC, Zerhouni EA, Judd RM, et al. Prognostic significance of microvascular obstruction by magnetic resonance imaging in patients with acute myocardial infarction. Circulation 1998;97:765-772.

19. Muehling OM, Wilke NM, Panse P, et al. Reduced myocardial perfusion reserve and transmural perfusion gradient in heart transplant arteriopathy assessed by magnetic resonance imaging. J Am Coll Cardiol 2003;42:1054-1060.

20. Al-Saadi N, Nagel E, Gross M, et al. Noninvasive detection of myocardial ischemia from perfusion reserve based on cardiovascular magnetic resonance. Circulation 2000;101:1379-1383.

21. Fenchel M, Franow A, Stauder NI, et al. Myocardial perfusion after angioplasty in patients suspected of having single-vessel coronary artery disease: improvement detected at rest-stress first-pass perfuion MR imaging-initial experience. Radiology 2005;237:67-74.

22. Panting JR, Gatehouse PD, Yang GZ, et al. Abnormal subendocardial perfusion in cardiac syndrome X detected by cardiovascular magnetic resonance imaging. NEJM 2002;346:1948-1953.

23. Kim WY, Danias PG, Stuber M, et al. Coronary magnetic resonance angiography for the detection of coronary stenoses. NEJM 2001;345:1863-1869.

24. Rickers C, Kraitchman D, Fischer G, et al. Cardiovascular interventional MR imaging: a new road for therapy and repair in the heart. Magn Reson Imaging Clin N Am 2005;13:465-479.

25. Araoz PA,Glockner JF, McGee KP, et al. 3 Tesla MR imaging provides improved contrast in first-pass myocardial perfusion imaging over a range of gadolinium doses. J Cardiovasc Magn Reson 2005;7:559-564.