Published online before print April 26, 2006, 10.1148/radiol.2393050228
(Radiology 2006;239:856-862.)
© RSNA, 2006
Cine Delayed-Enhancement MR Imaging of the Heart: Initial Experience1
Randolph M. Setser, DSc,
Jae K. Kim, MD, PhD,
Yiu Cho Chung, PhD,
Kevin Chen, MD,
Arthur E. Stillman, MD, PhD,
Ralf Loeffler, PhD,
Orlando P. Simonetti, PhD,
Joan A. Weaver, RT,
Michael L. Lieber, MS and
Richard D. White, MD
1 From the Section of Cardiovascular Imaging, Division of Radiology (R.M.S., J.K.K., K.C., A.E.S., J.A.W., R.D.W.), and Department of Biostatistics and Epidemiology (M.L.L.), the Cleveland Clinic Foundation, 9500 Euclid Ave, Desk HB6, Cleveland, OH 44195; Siemens Medical Solutions, Chicago, Ill (Y.C.C., O.P.S.); and Siemens Medical Solutions, Erlangen, Germany (R.L.). Received February 11, 2005; revision requested April 11; revision received June 30; accepted July 21; final version accepted September 1.
Address correspondence to R.M.S. (e-mail: setserr{at}ccf.org).
 |
ABSTRACT
|
|---|
This study was performed by using an institutional review board–approved protocol, with waived informed consent and HIPAA compliance. The purpose of this study was to preliminarily evaluate a cine delayed-enhancement (DE) pulse sequence for depiction of wall motion and myocardial scar extent during a single acquisition. The technique is based on inversion-recovery single-shot balanced steady-state free precession magnetic resonance imaging. Cine DE images were acquired in 26 patients (18 men, eight women; age range, 25–84 years; mean age, 61 years ± 13 [standard deviation]). Image contrast was consistent throughout each series. Overall (ie, with both readers' scores averaged), the cine DE imaging–depicted wall motion was scored correctly in 71% of myocardial segments. Scar extent was scored correctly in 76% of segments; in no patient was scarring missed. Cine DE imaging is a promising technique for simultaneous visualization of wall motion and myocardial scar extent.
© RSNA, 2006
|
INTRODUCTION
|
|---|
Preserved left ventricular (LV) function and viable LV myocardium are important positive prognostic indicators in patients with coronary artery disease (1). Magnetic resonance (MR) imaging is unique in its capability to enable accurate characterization of segmental LV dysfunction and direct visualization of nonviable myocardium during a single noninvasive examination. Previous study results have demonstrated that the extent of nonviable tissue in a myocardial segment can be used to predict recovery of function after revascularization (2,3).
Delayed-enhancement (DE) MR imaging is now considered the reference standard for direct visualization of nonviable LV myocardium. The technique is both accurate and robust (4,5) and enables visualization of small nontransmural infarcts that is not possible with other imaging modalities (6). In addition, a single-shot DE technique has been implemented (7) and shown to be useful in patients who are unable to suspend respiration sufficiently for completion of a segmented acquisition (8,9).
With use of existing techniques, viability and function data must be acquired separately, so multiple image acquisitions, which lengthen the examination time and can potentially result in spatial misregistration between images, are required. Furthermore, image interpretation can be hampered by the need to mentally integrate the results from these disparate sources. However, a technique that enabled simultaneous visualization of both viability and wall motion could improve image interpretation and thus enable physicians to not only discriminate viable from nonviable myocardium but also distinguish normal from hibernating (ie, viable but dysfunctional) myocardium. Thus, the purpose of this study was to preliminarily evaluate a cine DE pulse sequence for depiction of wall motion and myocardial scar extent during a single acquisition.
|
MATERIALS AND METHODS
|
|---|
The authors who were not employees of Siemens Medical Solutions (Erlangen, Germany), the manufacturer of the MR magnet used in this study, had control of the data and the information submitted for publication.
Patients and Imaging
This study was performed by using an active ongoing protocol—that is, it is reviewed yearly and approved by our institutional review board—and with waived informed patient consent, because the technology is associated with a nonsubstantial risk, as defined by Food and Drug Administration regulations. This study was Health Insurance Portability and Accountability Act compliant.
Cine DE images were acquired in 26 patients (18 men, eight women; age range, 25–84 years; mean age, 61 years ± 13 [standard deviation]) who underwent clinical assessment of myocardial viability from June through November 2004. Cine DE images were acquired in only cooperative patients who were able to adequately suspend respiration. In the first 15 patients, the image acquisition planes required to best visualize myocardial scarring in individual patients were selected; typically, only long-axis views of the LV were acquired. In the remaining 11 patients, midventricular (n = 10) or basal (n = 1) LV short-axis images also were acquired for comparisons with other image types.
Pulse Sequence
The described cine DE technique is based on inversion-recovery single-shot balanced steady-state free precession MR imaging (7). The timing of the pulse sequence components (Fig 1) is such that each image frame of the electrocardiographically triggered cine series is acquired during a separate R-R interval by using a constant TI. To create a cine series, the trigger delay is varied between image frames (Fig 2) to result in a series of single-shot images, each acquired during a different phase of the cardiac cycle. The first image is acquired at a trigger delay of 0 msec—that is, the trigger time equals the TI approximately 200–300 msec into systole. For the acquisition of subsequent images, with the trigger time longer than the TI, the trigger time is defined as the sum of the trigger delay plus the TI (Fig 2). However, to acquire early systolic images—that is, those acquired with a trigger time shorter than the TI—the trigger delay is set so that the image acquisition occurs during the subsequent heart beat (Fig 2).
View larger version (17K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1: Illustration of timing of events in the cine DE pulse sequence. Each image frame of the cine series is acquired by using a fixed inversion time (TI), which is set to null the signal from viable myocardium, as illustrated. However, the trigger delay (TDEL) is varied between image frames. Acq = acquisition, IR = inversion recovery.
|
|
View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2: Illustration of how trigger times are defined by using the cine DE sequence. For all images, the trigger delay (TDEL) for the inversion prepulse is defined relative to the first detected electrocardiographic R peak (top row). For images obtained with a trigger time longer than or equal to the sequence TI (Trigger Time1, middle row), the trigger time is also defined relative to the first detected electrocardiographic R peak. However, for images obtained with a trigger time shorter than the sequence TI (Trigger Time2, bottom row), the trigger time is defined relative to the second detected electrocardiographic R peak. Acq = acquisition.
|
|
The cine DE sequence was implemented by using a Magnetom Sonata 1.5-T MR imaging unit (Siemens Medical Solutions) and the following acquisition parameters: 2.5/1.1 (repetition time msec/echo time msec), a 50° flip angle, a bandwidth of 1090 Hz/pixel, a 380-mm field of view, a 75% rectangular field of view, an acquisition matrix of 192 x 115 (frequency x phase), and a temporal resolution of 287 msec. By default, 15 image frames were acquired in each cine DE series; the images were temporally spaced a mean of every 62 msec ± 9 (standard deviation). Images were acquired during a single breath hold that lasted 18–25 seconds, depending on the heart rate, approximately 10–20 minutes after the intravenous injection of 40 mL of 0.5 mmol/mL gadopentetate dimeglumine (Magnevist; Berlex Imaging, Wayne, NJ). The optimal TI needed to null viable myocardium was determined empirically for each patient and ranged between 250 and 325 msec.
As part of the clinical protocol, standard DE MR images were acquired in each patient by using an inversion-recovery spoiled turbo gradient-echo pulse sequence (10) with the following parameters: 8/4, a 30° flip angle, a bandwidth of 140 Hz/pixel, 23 k-space lines acquired every other R-R interval, a 300–360-mm field of view, an 80%–100% rectangular field of view, and an initial matrix of 256 x 256. Three LV long-axis views (horizontal long axis, vertical long axis, and LV outflow tract) and three LV short-axis views (base, apex, and middle ventricle) were acquired at mid-diastole in each patient. Images were acquired during successive 6–10-second breath holds. In each patient, the TI (range, 225–275 msec) was optimized to null viable myocardium. The time that elapsed between the intravenous injection of gadopentetate dimeglumine and the beginning of standard DE imaging was approximately 15–20 minutes.
Cine-loop MR images also were acquired, at short-axis levels that covered the LV, in each patient by using a balanced steady-state free precession sequence with retrospective electrocardiographic gating. Imaging parameters included 65–70/1.50–1.65, a 49°–65° flip angle, a section thickness of 6 or 10 mm, an x-axis field of view of 263–360 mm, a y-axis field of view of 300–360 mm, an initial matrix of 256 x 256, and a temporal resolution 25–43 msec. The breath-hold duration was 10–15 seconds, depending on the heart rate.
Image Analyses
Using commercially available image analysis software (Syngo 3D; Siemens Medical Solutions), an author (R.M.S., 10 years experience in cardiac imaging) manually drew regions of interest in the viable myocardium, which was dark; in the nonviable myocardium, which was highly enhanced; and in the LV blood pool, which was partially enhanced, in each image frame in the short-axis cine DE series. The regions were selected, at the operator's discretion, to include as much tissue as possible without incorporating the other tissue types. The average image pixel value for each region of interest was determined and then used to compute the relative contrast between the tissue types—specifically, between viable and nonviable tissue and between nonviable tissue and the LV blood pool.
Two readers (R.D.W, 21 years experience; A.E.S., 16 years experience) independently evaluated both the wall motion and the scar extent, first by using the cine DE images obtained in each patient. For the purposes of this study, each reader separately evaluated segmental wall motion on corresponding cine-loop MR images and evaluated myocardial scar extent on corresponding standard DE images, approximately 30 minutes after the cine DE image analysis. In all instances, the images were presented to each reader in random order and without patient identifiers. Each reader was blinded to the image findings of the other reader. With each image type, three myocardial segments—corresponding to the assumed left anterior descending artery, left circumflex artery, and right coronary artery distributions—were evaluated on each short-axis section (11). To this end, the LV was subjectively divided into three equiangular segments beginning at the midinterventricular septum. A scar extent score was assigned to each region by using a three-point scale, on which a score of 0 meant no scarring, 1 meant 1%–50% scarring, and 2 meant 51%–100% scarring. A wall motion score also was assigned to each region, again by using a three-point scale: A score of 0 meant normal, 1 meant hypokinetic, and 2 meant akinetic or dyskinetic motion.
Statistical Analyses
An author (M.L.L., 9 years experience) used weighted 
statistics calculated with statistical software (SAS, version 8; SAS Institute, Cary, NC) to quantify the level of agreement between the scar extent scores derived by using the cine DE images and those derived by using the standard DE images (12,13). Similarly, the level of agreement between the wall motion scores derived by using the cine DE images and those derived by using the cine-loop MR images also was assessed by using weighted statistics. Weighted statistics were used to account for the degree of disagreement between the techniques. In these analyses, each myocardial segment was considered to be an independent observation. Although the segments in a patient may have been correlated with respect to wall motion or scar extent, there is no reason to believe that the difference in wall motion or scar extent measurements (in the same set of segments) between two modalities was correlated among the segments within a patient. Thus, statistical methods to adjust for data clustering were not used.
|
RESULTS
|
|---|
In four (15%) of the 26 patients, no scar was visible on the long-axis or short-axis images; this was verified at standard DE MR imaging. In 21 (95%) of the remaining 22 patients, the appearance of scarring on the cine DE images (Figs 3–5) was consistent with that on the standard DE images. In the one patient with discrepant findings (Fig 5), the appearance of a small epicardial scar was not visualized well on the standard DE image, but it could be seen moving in and out of plane on the cine DE image.
View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3: Top: End-diastolic (ED), end-systolic (ES), and mid-diastolic short-axis midventricular cine DE MR images (2.5/1.1/230 [repetition time msec/echo time msec/TI msec]) in 84-year-old man with chronic ischemic heart disease show a high-enhancing nontransmural infarct (arrowheads) and corresponding systolic dysfunction in the septum and anterior walls. Relatively normal wall thickening is observed in the lateral wall at this anatomic level. Bottom: Corresponding end-diastolic and end-systolic cine-loop MR images (66/1.7) and mid-diastolic standard DE image (8/4/250).
|
|
View larger version (128K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4: Top: End-diastolic (ED), end-systolic (ES), and mid-diastolic horizontal long-axis cine DE MR images (2.5/1.1/250) in 75-year-old man with chronic ischemic heart disease show a transmural infarct (arrowhead) and akinetic wall motion at the apex. Wall motion is hypokinetic in the septum and lateral free wall. Bottom: Corresponding end-diastolic and end-systolic cine-loop MR images (65/1.6) and mid-diastolic standard DE image (8/4/250).
|
|
View larger version (108K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5: Top: End-diastolic (ED), end-systolic (ES), and mid-diastolic basal short-axis cine DE MR images (2.5/1.1/250) in 25-year-old man with mild focal epicardial scarring in the basal-inferolateral region (arrowheads) but no associated systolic dysfunction. Scar size varied throughout the cine DE imaging series, presumably because of through-plane myocardial motion at this level. Normal wall thickening was observed in the remainder of the LV. Bottom: Corresponding end-diastolic and end-systolic cine-loop MR images (65/1.6) and mid-diastolic standard DE image (8/4/250). Note that scarring (arrowheads) is visible on the cine-loop images but less apparent on the standard DE image at this anatomic level.
|
|
Image contrast was consistent throughout each cine DE series. The mean ratio of nonviable myocardium signal intensity–to–viable myocardium signal intensity was 7.0 ± 3.0 (standard deviation) (range, 2.5–17.1). The mean ratio of nonviable myocardium signal intensity–to–LV blood pool signal intensity was 1.6 ± 0.7 (range, 0.9–3.9).
Analysis of Segmental Scar Extent
According to reader 1, in the 11 patients (with total of 33 segments) with corresponding cine DE and standard DE short-axis images, myocardial scarring was present in 23 (70%) cine DE imaging–depicted segments and in 21 (64%) standard DE imaging–depicted segments. Furthermore, all 11 patients (100%) with scarring in at least one standard DE imaging–depicted segment also had scarring in at least one cine DE imaging–depicted segment, although the scar extent score sometimes differed between the image types.
Reader 1 correctly scored the scar extent in 26 (79%) of the 33 segments by using cine DE images—that is, the cine DE and standard DE imaging results matched. The cine DE imaging–derived scar extent score differed from the standard DE imaging–derived score by +1 (ie, was overestimated by one point) for four (12%) segments and by –1 (ie, was underestimated by one point) for three (9%) segments, but it differed by ±2 for no segments. The weighted coefficient for reader 1 was 0.75.
Reader 2 judged myocardial scarring to be present in 21 (64%) of the 33 cine DE imaging–depicted segments and in 25 (76%) of the standard DE imaging–depicted segments. According to reader 2, all 11 patients (100%) with scarring in at least one standard DE imaging–depicted segment also had scarring in at least one cine DE imaging–depicted segment, although the scar extent score sometimes differed. Reader 2 correctly scored the scar extent in 24 (73%) segments by using the cine DE images. The cine DE imaging–derived scar extent score differed from the standard DE imaging–derived score by +1 for three segments (9%) and by –1 for six segments (18%), but it did not differ by ±2 for any segment. The weighted coefficient for reader 2 was 0.68.
Reader 1 missed the scarring in one (5%) segment out of a total of 21 segments containing scar tissue—that is, the segment had a scar extent score of 0 at cine DE imaging but a scar extent score of 1 at standard DE imaging. Reader 2 missed the scarring in five (20%) segments out of a total of 25 segments containing scar tissue. In all cases, the missed segments contained small subendocardial regions of scarring; however, in no patient was the presence of nonviable myocardium missed altogether.
Analysis of Segmental Wall Motion
Reader 1 correctly scored the segmental wall motion in 24 (73%) of the 33 segments by using the cine DE images—that is, the cine-loop and cine DE imaging results matched. The cine DE imaging–derived wall motion score differed from the cine-loop MR imaging–derived score by +1 for five (15%) segments and by –1 for four (12%) segments, but it differed by ±2 for no segments. The resulting weighted coefficient for reader 1 was 0.64.
Reader 2 correctly scored the segmental wall motion in 23 (70%) of the 33 segments by using the cine DE images. The cine DE imaging–derived wall motion score differed from the cine-loop MR imaging–derived score by +1 for four (12%) segments and by –1 for six (18%) segments, but it differed by ±2 for no segments. The weighted coefficient for reader 2 was 0.66.
Reader 1 missed the abnormal wall motion in three (12%) segments out of a total of 25 segments with abnormal wall motion—that is, the segments had a wall motion score of 0 at cine DE imaging but a score of greater than 0 at cine-loop imaging. Reader 2 missed the abnormal wall motion in six (24%) segments.
|
DISCUSSION
|
|---|
In this study, we evaluated an imaging technique that enables the simultaneous visualization of LV wall motion and myocardial scar extent. Preliminary results indicate that this technique potentially could facilitate improvements in the noninvasive assessment of the effects of coronary artery disease. In nontransmural infarcts, cine DE imaging could enable physicians to more easily discriminate nonviable myocardium from myocardium that is viable yet dysfunctional. For example, the status of an epicardial rim of viable myocardium adjacent to a subendocardial infarct potentially could be better assessed by using this technique rather than assessing function and scar extent separately during two separate image acquisitions. Furthermore, in some cases, myocardial scarring could be better visualized during a cardiac phase other than mid-diastole, the phase during which DE images are typically acquired.
Analysis of Segmental Scar Extent
DE MR imaging with inversion-recovery single-shot balanced steady-state free precession has been shown to be useful in patients who are unable to adequately suspend respiration (7,8). However, as expected, the reduced breath-hold duration that is possible with the single-shot technique comes at the expense of lower spatial and temporal resolution compared with the spatial and temporal resolution achieved with segmented k-space acquisition. Hence, the image quality, as well as the conspicuity of nonviable myocardium, is typically decreased on single-shot images (8,9).
Both readers in the current study observed fairly good agreement (13) between the scar extent depicted on the cine DE images and that depicted on the standard DE images. The fact that subendocardial scarring was not observed on some cine DE images can be attributed to the lower spatial and temporal resolution of this sequence compared with that of segmented acquisition sequences.
Analysis of Segmental Wall Motion
The utility of real-time cine MR imaging for evaluation of LV function has been reported in several previously published studies (14–16). Real-time cine imaging seems to be the technique most analogous to cine DE imaging for comparison of wall motion results. In one study (15), wall motion was scored qualitatively by using a system similar to that used in the current study. In that study, with a temporal resolution of 87 msec per image, the real-time results were found to agree with the conventional cine imaging results for 691 (95%) of 724 myocardial segments; the value was 0.896.
Similar to the scar extent analysis results described earlier, fairly good agreement (13) between the cine DE imaging– and cine-loop imaging–depicted wall motion was observed by both readers. We believe that some dysfunctional segments were missed on the cine DE images because of the reduced temporal resolution. In addition, because systolic images are acquired last and during a cardiac cycle subsequent to the detected R wave, the cine DE sequence is susceptible to beat-to-beat variations in the R-R interval. This could lead to systole being undersampled and thus make evaluation of systolic function difficult in some cases.
Study Limitations
This study had the following limitations: (a) Only one image section was acquired and analyzed in each patient; (b) the analyses of wall thickening and scar extent were semiquantitative; (c) the segmental analysis was limited to the assessment of three segments per level instead of six segments per level, as recommended by the American Heart Association (11); and (d) the time between the analyses of the different image types was short (approximately 30 minutes).
In conclusion, cine DE MR imaging is a promising technique for the simultaneous visualization of wall motion and myocardial scar extent, and the use of this combination analysis could potentially improve the interpretation of images obtained in patients with ischemic heart disease. However, the current temporal resolution of this sequence is a limitation. Efforts to improve the temporal resolution by incorporating parallel imaging and/or segmented k-space schemes are currently underway.
|
ADVANCES IN KNOWLEDGE
|
|---|
- An acquisition technique for the simultaneous imaging of myocardial scar extent and cardiac function has been developed.
- In this study, the authors performed an initial clinical validation of a cine DE MR imaging technique.
|
FOOTNOTES
|
|---|
Abbreviations: DE = delayed enhancement • LV = left ventricle • TI = inversion time
See Materials and Methods for pertinent disclosures.
Author contributions: Guarantor of integrity of entire study, R.M.S.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, R.M.S., Y.C.C.; clinical studies, R.M.S., J.K.K., A.E.S., R.L., J.A.W., R.D.W.; statistical analysis, R.M.S., M.L.L.; and manuscript editing, all authors
|
References
|
|---|
- Allman KC, Shaw LJ, Hachamovitch R, Udelson JE. Myocardial viability testing and impact of revascularization on prognosis in patients with coronary artery disease and left ventricular dysfunction: a meta-analysis. J Am Coll Cardiol 2002;39:1151–1158.[Abstract/Free Full Text]
- Kim RJ, Fieno DS, Parrish TB, et al. Relationship of MRI delayed contrast enhancement to irreversible injury, infarct age, and contractile function. Circulation 1999;100:1992–2002.[Abstract/Free Full Text]
- Schvartzman PR, Srichai MB, Grimm RA, et al. Nonstress delayed-enhancement magnetic resonance imaging of the myocardium predicts improvement of function after revascularization for chronic ischemic heart disease with left ventricular dysfunction. Am Heart J 2003;146:535–541.[CrossRef][Medline]
- Fieno DS, Kim RJ, Chen EL, Lomasney JW, Klocke FJ, Judd RM. Contrast-enhanced magnetic resonance imaging of myocardium at risk: distinction between reversible and irreversible injury throughout infarct healing. J Am Coll Cardiol 2000;36:1985–1991.[Abstract/Free Full Text]
- Mahrholdt H, Wagner A, Holly TA, et al. Reproducibility of chronic infarct size measurement by contrast-enhanced magnetic resonance imaging. Circulation 2002;106:2322–2327.[Abstract/Free Full Text]
- Klein C, Nekoll SG, Bengel FM, et al. Assessment of myocardial viability with contrast-enhanced magnetic resonance imaging: comparison with positron emission tomography. Circulation 2002;105:162–167.[Abstract/Free Full Text]
- Chung YC, Vargas J, Simonetti O, Kim R, Judd R. Infarct imaging in a single heart beat [abstract]. J Cardiovasc Magn Reson 2002;4:12.
- Carr J, Pereles F, McCarthy R, et al. Delayed contrast-enhanced imaging of left ventricular myocardial scarring using single-shot inversion recovery (IR) TrueFISP [abstract]. In: Proceedings of the Eleventh Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 2003; 1612.
- Li W, Li BSY, Polzin JA, Mai VM, Prasad PV, Edelman RR. Myocardial delayed enhancement imaging using inversion recovery single-shot steady-state free precession: initial experience. J Magn Reson Imaging 2004;20:327–330.[CrossRef][Medline]
- Simonetti OP, Kim RJ, Fieno DS, et al. An improved MR imaging technique for the visualization of myocardial infarction. Radiology 2001;218:215–223.[Abstract/Free Full Text]
- Cerqueira MD, Weissman NJ, Dilsizian V, et al. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart. Circulation 2002;105:539–542.[Free Full Text]
- Kundel HL, Polansky M. Measurement of observer agreement. Radiology 2003;228:303–308.[Abstract/Free Full Text]
- Altman DG. Practical statistics for medical research. London, England: Chapman & Hall, 1991; 404.
- Setser RM, Fischer SE, Lorenz CH. Quantification of left ventricular function with magnetic resonance images acquired in real time. J Magn Reson Imaging 2000;12:430–438.[CrossRef][Medline]
- Plein S, Smith WHT, Ridgway JP, et al. Qualitative and quantitative analysis of regional left ventricular wall dynamics using real-time magnetic resonance imaging: comparison with conventional breath-hold gradient echo acquisition in volunteers and patients. J Magn Reson Imaging 2001;14:23–30.[CrossRef][Medline]
- Lee VS, Resnick D, Bundy JM, Simonetti OP, Lee P, Weinreb JC. Cardiac function: MR evaluation in one breath hold with real-time true fast imaging with steady-state precession. Radiology 2002;222:835–842.[Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
