| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Experimental Studies |
1 From the Laboratory of Cardiac Energetics, National Heart, Lung, and Blood Institute, National Institutes of Health, Department of Health and Human Services, 10 Center Dr, Bldg 10, Rm B1D416, MSC 1061, Bethesda, MD 20892-1061. Received February 11, 2002; revision requested March 11; revision received May 20; accepted July 24. Address correspondence to A.E.A. (e-mail: araia@nih.gov).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Twelve dogs were imaged on days 1 and 8 after transient 90-minute coronary artery occlusion. A magnetization transfer contrast with echo-train readout (MTET) MR sequence was performed before and 30 minutes after gadolinium contrast enhancement. Ex vivo analysis consisted of MR imaging, microsphere blood flow analysis, and triphenyltetrazolium chloride (TTC) staining. A paired two-tailed t test was used to compare wall thickening from day 1 to day 8. Linear regression and Bland-Altman analyses were used to compare infarct size depicted with MTET imaging with that seen on TTC-stained tissue.
RESULTS: Severe wall motion abnormalities were detected in all dogs. At TTC analysis, seven dogs had evidence of myocardial infarction and five had evidence of stunned myocardium. The mean percentages of left ventricular wall thickening in infarcted, stunned, and remote myocardial regions were 2% ± 4 (SD), 4% ± 8, and 33% ± 5, respectively. Wall thickening did not improve in the infarcted zones, but it improved to nearly normal levels in the stunned region 1 week after induced occlusion (mean, 40% ± 8; P < .02). MTET images clearly depicted infarcted myocardium as brighter than both the normal and stunned myocardial regions but darker than the blood pool. In vivo MTET infarct volume correlated with ex vivo TTC analysis data (y = 1.01x + 0.00, R = 0.98, standard error of the estimate = 0.019).
CONCLUSION: One day after myocardial ischemia, MTET during one MR imaging examination enabled simultaneous differentiation of infarcted, stunned, and normal myocardial regions on the basis of gadolinium enhancement and regional function.
Supplemental material: radiology.rsnajnls.org/cgi/content/full/2263012196/DC1
© RSNA, 2003
Index terms: Animals • Experimental study • Heart, experimental studies • Magnetic resonance (MR), magnetization transfer contrast, 511.121417 • Myocardium, infarction, 511.77 • Myocardium, MR, 511.121411, 511.121412, 511.121413, 511.121416, 511.121417, 511.12143
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Magnetization transfer (MT) is a term used to describe the exchange of magnetization between water and macromolecular protons. MT is typically detected by means of applying a narrow-band radio-frequency pulse to the macromolecular protons off resonance prior to image acquisition. In the presence of MT, the signal intensity of water is decreased in rough proportion to the macromolecular proton–to-water exchange rate and the T1 relaxation of the water resonance. As such, MT is dependent on both the macromolecule exchange rate and the water proton T1. Investigators have used this T1 dependence of MT to enhance the gadolinium contrast in the brain (7,8) with any underlying effect on the macromolecular proton–to-water exchange rate. Because of the high macromolecular content of heart muscle, macromolecular-water proton MT is substantial in the heart, whereas the relatively low macromolecular content of blood leads to excellent muscle-to-blood contrast (9).
We hypothesized that with this strong MT background in the heart, MT could enable the detection of gadolinium retention in infarcted myocardium by way of the change in water T1 associated with the contrast agent. In addition, we proposed that MT-generated blood-to-tissue contrast might prove to be useful in resolving the blood cavity from infarcted myocardium in patients with endocardial lesions. It is important to note that the T1 contrast generated with MT can be achieved by using off-resonance radio-frequency radiation without increasing the imaging flip angle. Higher flip angle leads to flow and motion artifacts in the heart. Thus, the purpose of this study was to simultaneously differentiate stunned, infarcted, and normal myocardial regions in dogs by using gadolinium-enhanced cine MR imaging combined with MT contrast.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
One week later, the animals were subjected to 90 minutes of left anterior descending coronary artery occlusion followed by reperfusion; sedation was induced with intramuscularly administered butorphanol tartrate (0.4 mg/kg), and pain was controlled with intravenously administered oxymorphone (0.2 mg/kg) and 2.5% thiopental sodium (1 mL per 3 kg). Pain following occlusion was managed with intravenous doses of buprenorphine hydrochloride (6 mg/kg). Ventricular arrhythmias were treated with lidocaine and cardioversion, as indicated.
MR Imaging Examinations
MR imaging to evaluate global and regional left ventricular function, myocardial perfusion, and delayed gadolinium hyperenhancement was performed 24 hours after induced reperfusion (day 1) and immediately before sacrifice (day 8). The MR imaging evaluation consisted of a closed-chest procedure with intravenous oxymorphone- and pentothal-induced anesthesia. When the dogs reached the stage 2 anesthetic state, they were intubated and ventilated. Vecuronium (0.1 mg/kg) was administered intravenously to ensure appropriate gating and ventilation during the breath-hold MR imaging examinations. Body temperature was maintained with heated water–circulating blankets. At completion of MR imaging, the effects of vecuronium were reversed with intravenous administration of neostigmine (0.001 mg/kg). When necessary, we reversed the effects of oxymorphone by intravenously administering naloxone (0.04 mg/kg).
MR imaging was performed at 1.5 T with a cardiovascular magnet system (model LX 8.25M4; GE Medical Systems, Milwaukee, Wis). Spins were excited through the radio-frequency body coil, and signals were received through a four-coil phased-array receive-only extremity coil. Conventional short-axis cine fast gradient-echo MR images (10) were acquired from the base to the apex of the heart. The imaging parameters were as follows: 11.6/6.5 (repetition time msec/echo time msec); in-plane resolution, 1.1 x 1.7 mm2; section thickness, 8 mm; sampling rate, 20 kHz; and flip angle, 15°. During the cardiac cycle, 20 cardiac images were reconstructed from images sampled at four views per segment, with a 46.4-msec acquisition window within the cardiac cycle.
MR imaging was performed before and approximately 30 minutes after the intravenous administration of 0.14 mmol/kg of gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ) to minimize the chances of overestimating infarct size (11). Three authors (A.H.A., F.H.E., R.S.B.) developed an experimental MR imaging sequence, MT contrast with echo-train readout (MTET) (Fig 1), by adding a 16-msec MT pulse to each repetition time of a fast gradient-echo sequence that was accelerated by short echo-planar readouts (12,13). The segmented echo-planar imaging parameters were as follows: echo time, 1.8 msec; in-plane resolution, 1.7 x 2.1 mm2; flip angle, 20°; echo train length, four; sampling rate, 62.5 kHz; field of view, 34.0 x 25.5 cm; section thickness, 8 mm; matrix size, 192 x 120, and one acquisition. The repetition time (28 msec) included 18 msec for MT and 10 msec for image acquisition. During the cardiac cycle, 20 images were reconstructed from data sampled at eight views per segment, with a 56-msec acquisition window within the cardiac cycle.
|
0.001%). Volumetric stimulated echo acquisition mode, or STEAM, was used to shim the left ventricular chamber and establish the correct transmit frequency. Different transmit frequency offsets, ranging between 300 and 600 Hz, were used for the MT pulse. A random-amplitude gradient-crushing pulse followed each MT pulse to scramble the residual transverse magnetization. We randomized the phase of the MT radio-frequency pulse to further facilitate this process. The peak radio-frequency magnetic induction field was 77.58 mG for a flip angle of 800°. To determine the effect of the MT preparation pulse, we conducted experiments with the MT flip angle at 0° (ie, MT off) and 800° (ie, MT on) both before and after contrast material administration. The contrast-to-noise ratio between the infarcted and remote zones was measured with the MT pulse off and with the MT pulse on, both before and after contrast material administration.
Postmortem inversion-recovery MR imaging parameters were as follows: 2,500/10; inversion time, 550 msec; in-plane resolution, 0.6 x 0.6 mm2; section thickness, 20 mm; sampling rate, 16 kHz; and spin-echo readout. Note that the section thickness was restricted by the 4-mm thickness of the tissue samples rather than by the imaging parameters per se.
Myocardial Tissue Processing
Authors (C.R.W., J.F.L., R.W., A.E.A.) measured myocardial perfusion by injecting approximately 10 million cross-linked polystyrene nonbiodegradable colored microspheres (Interactive Medical Technologies, Irvine, Calif) into the left atrial catheter for 15–30 seconds. The withdrawal of the blood reference was started before injection of the microspheres and was taken at 10 mL/min for 2 minutes through a 5-F femoral artery sheath and controlled with use of a withdrawal pump (Harvard Apparatus, South Natick, Mass). Myocardial blood flow was measured at baseline, 10 minutes into the left anterior descending coronary artery occlusion, during the first MR imaging examination (day 1), and during the second MR imaging examination (day 8).
To facilitate cutting, we suspended the ventricular myocardium in a 4% low-melting-point agarose solution, which was allowed to gel. The solidified substance was cut into 4-mm-thick slices by using a commercially available meat slicer (Globe Food Equipment, Dayton, Ohio). The slices were stained with triphenyltetrazolium chloride (TTC) and imaged ex vivo. Finally, each short-axis slice was divided into 32 segments (16 radial segments divided into endocardial and epicardial halves). Two neighboring 4-mm-thick slices were combined to produce an 8-mm-thick slice for quantitation of flow. Myocardial tissue and blood references were quantified by using a blood flow analysis system (Interactive Medical Technologies).
Authors (C.R.W., J.F.L., R.W., A.E.A.) evaluated the area of necrosis by using standard TTC staining methods (14). Planimetry was used (C.R.W., A.E.A.) to measure the left ventricular myocardium and the non–TTC-stained (ie, TTC-negative) region of the myocardium. The signal intensity of the myocardial regions was measured (C.R.W.) by using computer-assisted planimetry. The contrast-to-noise ratio was defined as the difference in signal intensity between two regions of interest divided by the SD of an artifact-free region of interest in air. No correction was applied to account for the nonnormal noise distribution on magnitude images. Ex vivo infarct volume imaging was performed on each 4-mm tissue slice because partial volume effects can contribute to discrepancies between the pathologic and imaging measurements of infarction.
Data and Statistical Analyses
Animals with infarction were defined as those showing evidence of infarction with ex vivo TTC analysis and inversion-recovery MR imaging (C.R.W., A.E.A.). Animals with stunned myocardium were defined as those showing no evidence of infarction with ex vivo TTC analysis but abnormal myocardial wall motion on day 1 of MR imaging. All animals were classified as having either infarcted or stunned myocardium; no animal had entirely normal myocardium after induced occlusion. Blinded to the ex vivo results, we assessed the in vivo MTET MR imaging results quantitatively and qualitatively to characterize the myocardial status of the animals. Authors (C.R.W., A.E.A.) used computer-assisted planimetry to measure left ventricular end-diastolic and end-systolic wall thicknesses. Wall thickening was defined as the percentage change in wall thickness from end diastole to end systole. Statistical comparisons were performed by using a paired two-tailed t test, an unpaired two-tailed t test, and simple linear regression. Bland-Altman plots were used to supplement the linear regression analyses. P > .05 indicated a nonsignificant difference. Unless otherwise noted, results are presented as means ± SDs.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
In Figure 3, a contrast-enhanced MTET MR image of myocardial infarction is compared with a conventional precontrast fast gradient-echo MR image, an ex vivo inversion-recovery MR image, and a TTC-stained pathologic section. On the fast gradient-echo image, the normal and infarcted myocardial regions have similar signal intensity. On the MTET image, the area of infarction is seen as a thick bright crescent in the anterior wall of the myocardium. Although the infarcted region appears bright compared with the surrounding signal intensity–suppressed myocardium, it is still easily distinguishable from the much brighter blood pool. All seven infarctions (in seven dogs) were visually identified on the in vivo MTET images obtained on days 1 and 8. None of the animals with only stunned myocardium showed myocardial hyperenhancement at any of the MR imaging examinations.
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Animal studies (6,15) have revealed a close correlation between gadolinium-enhanced MR imaging–based infarct size and pathologic measurements, but there are exceptions to this generalization. The infarct size measured with TTC staining correlated well with that measured on the basis of delayed gadolinium hyperenhancement in dogs (15), but this correlation measurement approach may lead to an overestimation of infarct size by about 12%. More recent work suggests that this correlation remains close from 1 day to at least 8 weeks after infarction in dogs (6). The correlation between infarct size determined with TTC staining and that determined with gadolinium-enhanced MR imaging is also good in rabbits (16). In a rat study (17) in which gadolinium-enhanced MR imaging led to an overestimation of infarct size, the findings may have been due to an inadequate reperfusion time and therefore a TTC analysis–based underestimation of infarct size. However, other rat infarction studies (18) have revealed a systematic overestimation of infarct size with TTC analysis. Certain infarct-avid gadolinium-based contrast agents may enable more accurate determination of infarct size (19). The timing of imaging after contrast material administration is also important for the accurate measurement of infarct size (11).
The results of human trials indicate that gadolinium-enhanced MR imaging can depict myocardial infarction (20–23) and enable differentiation between reperfused and nonreperfused infarction (24). Data obtained from humans have led to controversy as to whether viable myocardium may show delayed hyperenhancement. Thus, abnormal delayed hyperenhancement may not be synonymous with infarcted myocardium (25,26). Hyperenhanced segments have been responsive to the low-dose administration of dobutamine in the majority of cases in which contrast enhancement was limited to the subendocardial regions but also in about 40% of cases in which there was a transmural pattern of enhancement. Only the hyperenhancement associated with microvascular obstruction has seemed to indicate the presence of nonviable tissue with high certainty. These clinical findings highlight the importance of careful study validation and the need for additional studies of the pathophysiologic processes that modulate MR imaging contrast agents.
At least three mechanisms can lead to inaccurate determination of infarct size with gadolinium-based MR imaging: (a) A bright rim of blood in areas of stagnant flow may have an appearance similar that of a subendocardial infarction (27), (b) there may be a minimal difference in signal intensity between an infarcted region and the blood pool, and (c) a border zone of reversibly injured tissue may be enhanced by gadolinium-based contrast material. In addition, a lack of understanding of why gadolinium kinetics are different in infarcted versus normal myocardial regions adds to the uncertainty. The results of several studies of contrast material kinetics (15,16,28) and ion distribution in and around infarcted myocardium support the validity of these imaging features as markers of viability. The MTET approach provides the benefit of nearly equal contrast between the infarcted region and either the blood pool or the normal myocardium (Fig 5), which may minimize the chances of the first two types of errors (ie, bright rim of blood in areas of stagnant flow and minimal signal intensity difference between an infarcted region and the blood pool).
In gadodiamide-enhanced rat studies (29), stunned myocardium and normal myocardium had similar signal intensity, whereas infarcted myocardium enhanced as expected. In the study of Schwitter et al (18), however, TTC-negative tissue was enhanced in a few animals. Kim et al (6) induced 15-minute coronary occlusion in dogs to produce stunning without infarction and observed no hyperenhancement of the stunned myocardium. Our study data in dogs suggest that a 90-minute occlusion would cause a heart to have infarcted and/or stunned regions, presumably on the basis of collateral blood flow to these regions. Therefore, this animal model provided an excellent opportunity to evaluate the effects of stunning and infarction on gadolinium retention and myocardial function in the same model. A major finding of this study, that gadolinium retention did not occur in the stunned myocardium, indicates that abnormal gadolinium hyperenhancement has high specificity for the detection of infarcted myocardium in dogs.
Previously performed studies (30) in rats revealed that MT improves the contrast between normal and infarcted myocardium in the absence of gadolinium, apparently because of the increase in free water content (30). We were not able to reproduce this phenomenon in vivo in our current work. However, the previous study (30) was conducted with high (ie, 4.7-T) magnetic fields, which improved the MT effects by facilitating extended T1 and optimized the off-resonance irradiation conditions. In the current study, with only a 50% MT irradiation cycle, the use of segmented echo-planar imaging limited the magnitude of the MT effect and the signal-to-noise ratio on the MR images. The use of high–signal-to-noise imaging with more effective MT preirradiation may improve the capability of MT to depict infarcted regions without gadolinium enhancement.
Alternative methods such as spin-lock preparation can also be used to acquire simultaneous functional and viability data at MR imaging (31). In our experience, increasing the T1 weighting at conventional cine MR imaging by using a higher flip angle radio-frequency pulse may improve the differentiation between infarcted and normal myocardial regions, but it may also increase blood flow–related artifacts on images of the myocardium.
Practical application: Results of the studies described herein demonstrate that MTET can be used with MR imaging to detect postinfarction gadolinium retention. MTET involved the use of off-resonance radio-frequency radiation to generate T1 contrast by way of the MT mechanism and thus avoid the excessive excitation of water protons that occurs with increased on-resonance flip angles. The accuracy of infarct detection with MTET MR imaging was good compared with that of in vitro TTC staining. However, unlike static diastolic inversion-recovery sequences, MTET also generated dynamic cardiac wall motion information. The use of MTET enables the combined evaluation of cardiac dynamic and distribution gadolinium retention data in a single acquisition. This multiparameter approach has the potential to help reduce the time required to perform cardiac imaging examinations and to generate new information on the contractile consequences of a given infarction for better clinical evaluations.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, R.S.B., A.E.A.; study concepts, F.H.E., R.S.B., A.E.A., A.H.A.; study design, F.H.E., R.S.B., A.E.A., J.L.T., J.F.L., A.H.A., R.W.; literature research, R.S.B., C.R.W.; experimental studies, C.R.W., J.F.L., J.L.T.; data acquisition, C.R.W., A.H.A., J.L.T.; data analysis/interpretation, C.R.W., A.H.A., J.L.T., R.S.B., A.E.A., J.F.L., R.W.; statistical analysis, C.R.W., A.E.A.; manuscript preparation, C.R.W., A.E.A., R.S.B.; manuscript definition of intellectual content, C.R.W., A.E.A., A.H.A., R.S.B., J.F.L., R.W.; manuscript editing, revision/review, and final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. L. Raff, W. W. O'Neill, R. E. Gentry, A. Dulli, K. G. Bis, A. N. Shetty, and J. A. Goldstein Microvascular Obstruction and Myocardial Function after Acute Myocardial Infarction: Assessment by Using Contrast-enhanced Cine MR Imaging. Radiology, August 1, 2006; 240(2): 529 - 536. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Aletras, G. S. Tilak, A. Natanzon, L.-Y. Hsu, F. M. Gonzalez, R. F. Hoyt Jr, and A. E. Arai Retrospective Determination of the Area at Risk for Reperfused Acute Myocardial Infarction With T2-Weighted Cardiac Magnetic Resonance Imaging: Histopathological and Displacement Encoding With Stimulated Echoes (DENSE) Functional Validations Circulation, April 18, 2006; 113(15): 1865 - 1870. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-C. Toufektsian, Z. Yang, K. M. Prasad, L. Overbergh, S. I. Ramos, C. Mathieu, J. Linden, and B. A. French Stimulation of A2A-adenosine receptors after myocardial infarction suppresses inflammatory activation and attenuates contractile dysfunction in the remote left ventricle Am J Physiol Heart Circ Physiol, April 1, 2006; 290(4): H1410 - H1418. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Natanzon, A. H. Aletras, L.-Y. Hsu, and A. E. Arai Determining Canine Myocardial Area at Risk with Manganese-enhanced MR Imaging Radiology, September 1, 2005; 236(3): 859 - 866. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. R. Edelman Contrast-enhanced MR Imaging of the Heart: Overview of the Literature Radiology, September 1, 2004; 232(3): 653 - 668. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Muthupillai, S. D. Flamm, J. M. Wilson, R. I. Pettigrew, and W. T. Dixon Acute Myocardial Infarction: Tissue Characterization with T1{rho}-weighted MR Imaging--Initial Experience Radiology, August 1, 2004; 232(2): 606 - 610. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Shan, G. Constantine, M. Sivananthan, and S. D. Flamm Role of Cardiac Magnetic Resonance Imaging in the Assessment of Myocardial Viability Circulation, March 23, 2004; 109(11): 1328 - 1334. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
