Ann Thorac Surg 2005;79:1338-1343
© 2005 The Society of Thoracic Surgeons
Original article: Cardiovascular
Principal Strain Orientation in the Normal Human Left Ventricle
Brian P. Cupps, PhDa,
Benjamin J. Pomerantz, MDa,
Marc D. Krock, MDa,
Joseph Villard, MSEEa,
Joseph Rogers, MDb,
Nader Moazami, MDa,
Michael K. Pasque, MDa,*
a Department of Surgery, Division of Cardiothoracic Surgery, Washington University School of Medicine, St. Louis, Missouri, USA
b Department of Cardiovascular Medicine, Washington University School of Medicine, St. Louis, Missouri, USA
Accepted for publication May 3, 2004.
* Address reprint requests to Dr Pasque, Ste 3108, Queeny Tower, Barnes Jewish Hospital, One Barnes Jewish Hospital Plaza, St. Louis, MO 63110, USA
pasquem{at}msnotes.wustl.edu
 |
Abstract
|
|---|
BACKGROUND: Methods that can improve the accuracy of application of directed intervention in the treatment of coronary artery disease deserve investigation. Magnetic resonance imaging with tissue tagging allows for noninvasive, quantitative determination of regionally varying minimum principal strain. Because the directional vector of minimum principal strain has been shown to be sensitive to ischemic involvement, my colleagues and I sought to fully characterize the normal range of vector direction in the in vivo human left ventricle at rest and during inotropic stimulation.
METHODS: Tagged magnetic resonance imaging image sets were acquired in 20 healthy volunteers at rest and during dobutamine infusion. Strain was computed from the measured displacement data by using finite element software. Orientation of minimum principal strain was characterized by measuring the angle (principal strain angle) between the minimum principal strain vector and the local circumferential-longitudinal plane. Values of this angle were computed in 6 ventricular regions and globally.
RESULTS: Resting values of the principal strain angle were small in every region, confirming that maximal normal myocardial contraction occurs primarily in the circumferential-longitudinal plane. Angles were similar during dobutamine infusion. Comparisons between ventricular walls, both at rest and with dobutamine, revealed no marked regional differences in the principal strain angle.
CONCLUSIONS: The direction of maximal myocardial contraction is known to change with ischemic injury to the myocardium and can be a sensitive, regionally varying index of myocardial ischemia. The critical first step in the clinical application of this technology is to accurately characterize normal ranges of principal strain angles.
|
Introduction
|
|---|
Objective, regionally accurate characterization of myocardial viability and ischemic involvement before intervention in the setting of chronic ischemic coronary artery disease continues to be a diagnostic dilemma for clinicians. Traditional methods such as thallium scan imaging, positron emission tomography (PET), and stress echocardiography are currently used to make clinical decisions regarding both the presence of inducible myocardial ischemia and the reversibility of regional myocardial injury secondary to coronary artery disease. These methods are qualitative, by nature, in their assessment of the degree of ischemic involvement and exact regional localization of ischemia, and they are dependent on subjective observer variability [1]. Clearly, a method that could mathematically quantify regional and transmural ischemic involvement in functional terms would be a valuable clinical tool.
The current applications of magnetic resonance imaging (MRI) techniques in the evaluation of myocardial viability have largely involved the use of gadolinium for contrast-enhanced MRI cardiac imaging [2]. The various patterns of early hypoenhancement and delayed hyperenhancement after contrast injection have been variably correlated with myocardial viability. Kim and associates [3], for example, found correlation in myocardial segments that showed delayed hyperenhancement and no improvement in echocardiographically assessed contractility after revascularization. Gerber and associates [4] expressed limitations in the echocardiographic assessment of contractility and subsequently attempted to differentiate the significance of early hypoenhancement and delayed hyperenhancement by way of correlating these findings with contractility assessments that used radiofrequency tissue-tagging MRI. Their use of circumferential shortening strain for correlation with contractility in their receiver operator curves implies that radiofrequency tissue-tagging variables for contractility can be considered the "gold standard" for functionality evaluations. Additionally, the limited sensitivity of delayed hyperenhancement at 82% and a specificity of 64% for predicting improvement of dysfunctional segments argues that gadolinium-enhanced MRI still has ambiguities regarding its assessment of viability [4]. In light of this, MRI tissue-tagging variables seem to offer detailed information regarding myocardial function and strain for potential applications in assessing myocardial viability.
The inherent qualities of MRI combined with tissue tagging give these techniques a unique capability to quantify and accurately localize the effects of myocardial ischemia on myocardial contraction, independent of observer variability. By using MRI with tissue tagging, the magnetic properties of selective material points are altered to create tagged patterns within a deforming body, such as the heart. Differences in signal intensity between tagged and unaltered regions allow for tracking the motion of the underlying tissue. These tagged image sets provide information about local contraction and deformation of the myocardium, which can be used to obtain local strain values for different myocardial regions.
The purpose of this study was to accurately characterize the normal ranges of the principal strain angle (the angle between the minimum principal strain vector and the circumferential-longitudinal plane; see the Appendix) in healthy volunteers at rest and during dobutamine infusion. The minimum principal strain (
3) is the smallest eigenvalue of the strain tensor. In the heart during systole, it is associated with the magnitude of maximum contraction that occurs at any given material point. The associated eigenvector indicates the direction of maximal contraction at that point. The establishment of normal ranges of the principal strain angle is a critical first step in the utilization of the direction of maximal shortening as an indicator of ischemic involvement of the myocardium. This valuable clinical tool has the potential to mathematically quantify myocardial contractile performance and, thereby ischemic involvement, on a regional, transmural basis.
|
Material and Methods
|
|---|
Patient Characteristics
A group of 20 healthy volunteers (10 men and 10 women; age range, 20 to 38 years; mean age ± standard deviation, 27.7 ± 6.1 years) were studied. All volunteers had normal physical and electrocardiographic (ECG) findings and no history of heart disease. Additionally, 4 patients (3 men and 1 woman; age range, 58 to 67 years; mean age ± standard deviation, 63.2 ± 4.1 years) with documented coronary artery disease were also studied to explore the utility of these techniques in patients with ischemic cardiomyopathy. The study was approved by the Human Studies Committee at Washington University (St. Louis, MO), and all volunteers gave written, informed consent.
Imaging Protocol
Imaging was performed in a 1.5-T magnetic resonance scanner (Magnetom Vision; Siemens Medical Systems, Iselin, NJ) at rest and with a dobutamine (10 µg · kg–1 · min–1) infusion. A series of scout images was obtained to locate the heart and establish the long- and short-axis imaging planes. Subsequently, a set of parallel short-axis imaging planes (8 mm thick) were obtained at 8-mm intervals beginning at the level of the mitral valve and ending at a short-axis imaging plane that contained only apical myocardium and no left or right ventricular endocardium. An additional set of 4 long-axis imaging planes was obtained according to the following criteria: (1) orthogonal to the short-axis imaging planes, (2) intersecting the centroid of the left ventricle, and (3) oriented in a radial fashion with 45 degrees of separation between long-axis imaging planes.
Image Acquisition
Image acquisition was synchronized with real-time ECG at the time of the MRI scanning. The R wave of the ECG signal was used to activate the MRI scanner to commence scanning. During the actual image data acquisition, the subjects were instructed to hold their breath at mid expiration, and a series of images was acquired at 29-ms intervals until the approximate completion of the entire cardiac cycle at each imaging plane. Each cine sequence of images consisted of 12 to 18 images acquired over 25 to 30 seconds. The end-diastolic image was chosen as the first image in sequence, and the end-systolic image was determined by choosing the smallest ventricular size in sequence. Data acquisition time was approximately 60 minutes. For each selected imaging plane, a single-slice magnetic resonance–tagged image was collected with a sequence consisting of a spatial modulation of magnetization radiofrequency tissue-tagging preparation followed by a 2-dimensional fast low flip angle shot cine image acquisition. Imaging variables were a repetition time of 58 ms per cine segment, an echo time of 2.9 ms, an excitation angle (
) of 30 degrees, a slice thickness of 8 mm, and an acquisition matrix of 256 x 256. The field of view was set to 300 x 300 mm2 and 400 x 400 mm2 for the short- and long-axis images, respectively.
Measurements of 3 direction were obtained from a finite element model of the mid ventricle by using a previously described and validated method [5, 6]. Average values of the principal strain angle were computed over the anteroseptal, anterior, anterolateral, posterolateral, posterior, and posteroseptal walls at rest and during dobutamine infusion. Classification and nomenclature for these elements followed the recommendations of the American Society of Echocardiography Committee on Standards [7]. A global average was computed from the 6 regional average values. These angle measurements can also be presented graphically. A principal strain angle contour map from a healthy volunteer is presented in Figure 1A.
View larger version (38K):
[in this window]
[in a new window]
|
Fig 1. (A) Principal strain angle color contour map from a healthy volunteer. The anterior wall of the ventricle is oriented toward the front of the model. (B) Principal strain angle color contour map of a 62-year-old male patient with known coronary artery disease. The resting principal strain angle values were abnormal in both the anterior and anterior septal regions. Regions of blue indicate a principal strain angle of less than 30 degrees. Green areas represent angle values between 30 and 60 degrees, and red regions represent angle values greater than 60 degrees.
|
|
Statistical Analysis
Statistical analysis was performed with SPSS software (SPSS Inc, Chicago, IL). Comparisons between the same ventricular regions at rest and during dobutamine infusion were performed with a paired t test. Comparisons between the 6 ventricular regions at rest and with dobutamine were performed by using analysis of variance; post hoc testing was performed with Fisher's least significant difference test. In all cases, a p value less than 0.05 was considered significant.
|
Results
|
|---|
Average principal strain angles from the 20 healthy volunteers at rest and during dobutamine administration are presented in Figure 2. At rest, these angles were small, ranging from 8 to 10 degrees, with standard deviations ranging from 2 to 4 degrees. Dobutamine infusion did not significantly alter these angle measurements in any region (p > 0.5 for all regions). Comparisons between the 6 ventricular walls, both at rest (p = 0.369) and with dobutamine (p = 0.104), revealed no significant regional differences in the principal strain angle.
These data suggest that in healthy myocardium, the minimum principal strain is oriented approximately in planes (circumferential-longitudinal) that are tangential to the ventricular walls both at rest and during dobutamine administration. To illustrate the potential clinical application of these findings, my colleagues and I evaluated 4 ischemic cardiomyopathic patients with abnormal principal strain angles by using other imaging modalities (PET, thallium scan imaging, echocardiography, and angiogram). As shown in Table 1, segments for these patients found to have abnormal angles with dobutamine were found to be nonviable by PET and other modalities. A resting principal strain angle contour map for one of these patients is given in Figure 1B.
View this table:
[in this window]
[in a new window]
|
Table 1. Principal Strain Angles, PET, Thallium, Echocardiographic, and Angiographic Data From 4 Patients With Known Ischemic Coronary Artery Disease
|
|
|
Comment
|
|---|
One of the primary goals of myocardial imaging in patients with ischemic myocardium is to identify abnormally functioning myocardial segments that may return to normal after revascularization. Myocardium with chronically abnormal function that can be recruited to normal with revascularization is termed hibernating myocardium [8, 9].
Preoperative evaluation of myocardial viability with current imaging modalities continues to be suboptimal. Traditional methods used to determine myocardial viability, which include dobutamine echocardiography, thallium imaging, and PET, only qualitatively estimate regional viability. This study demonstrates a reliable, quantitative, regionally sensitive, nonsubjective method of determining preintervention myocardial viability by using principal strain angles as a measure of viability. Previous studies have shown that in normal myocardium, 3 is approximately oriented in planes defined by the local circumferential and longitudinal directions [10, 11]. Other research has demonstrated a reorientation of the principal strain directions during acute ischemia [12] and in infarcted myocardium [13]. More recently, Garot and colleagues [14] found in 2 dimensions that the angle between the minimum principal strain and the circumferential direction was increased in dysfunctional segments when compared with remote segments and was further augmented in akinetic segments.
Numerous comparisons have been made among "traditional" methods of myocardial evaluation, such as echocardiography, scans with nuclear tracers, and MRI [15–18]. Many of these methods rely on myocardial perfusion to determine viability. Hibernating myocardium may have diminished blood flow or even normal blood flow at rest. Nonviable myocardium may also have nearly normal blood flow, and this renders methods that use myocardial blood flow to determine viable myocardium less sensitive and less specific. In addition, myocytes within a hibernating region may undergo ultrastructural changes subsequent to an ischemic event [19]. These changes may lead to an alteration in metabolic activity, which again may lead to a false-negative or false-positive result. In addition to flow limitations, various regions of the heart can be difficult to visualize because of its location. In particular, the inferior wall may be problematic because of its approximation with the diaphragm [20]. Furthermore, investigators have reported the underestimation of hypoperfused myocardium with the use of technetium-labeled agent [21]. In another study, Doenst and Taegtmeyer [22] demonstrated an underestimation of glucose uptake by fluorodeoxyglucose (FDG) uptake upon reperfusion in rat hearts. In these experiments, there was a decrease in the uptake of FDG upon reperfusion in animals in the fed state and in which there were fatty acids present as a competing substrate for glucose. Although this study examined reperfusion in the acute setting, it may hold true in the chronically ischemic state. MRI imaging is able to perform contractile functional analysis rather than just providing data on the ability of the myocytes to take up substrate. In clinical outcomes, perfusion is important but can be misleading, whereas recovery of function remains the bottom line.
Baer and colleagues [16] demonstrated that dobutamine MRI was more sensitive and specific than transesophageal echocardiography for the diagnosis of residual myocardial viability after a myocardial infarction. In these studies, Baer and associates used PET scanning as the standard with which MRI and transesophageal echocardiography were compared. The predictive value of dobutamine echocardiography may also be decreased when myocardial perfusion is diminished because of the inherent difficulties in visualizing the entire myocardium and the gross nature and the subjectivity of the interpretation of the images produced by echocardiography.
The evaluation of myocardial function with MRI has been well characterized in both normal and pathologic states [10, 11, 13, 23–28]. Myocardial MRI images possess several advantages, most notably the high spatial resolution and sharp endocardial and epicardial border determination.
The results of this study quantify the 3-dimensional regional myocardial contractile response to dobutamine in the healthy in vivo human heart. These results strongly suggest that the principal strain angle is highly quantifiable, appropriately uniform, reproducible, and easily acquired in the clinical setting, thereby demonstrating its potential in this patient population. The results of this investigation can be used as a template to compare responses to dobutamine in various clinically important pathologic patient subsets, such as those associated with ischemic coronary artery disease, valvular disease, and myopathic disease. Quantification of the regional pathologic response may more accurately direct surgical or catheter-based intervention and be predictive of subsequent clinical response.
There are several advantages of using the principal strain angle to determine myocardial pathology: (1) it is objective and is not subject to operator bias, (2) it is noninvasive, (3) it is highly regionally and transmurally sensitive, (4) it does not require the use of radiation, (5) MRI imaging is able to image the entire ventricle, and (6) it is not blood flow dependent.
One limitation of this study was our use of only a single dose of dobutamine. However, strain maximizes at a 10 µg · kg–1 · min–1 dose [24, 29], and the imposition of the longer scanning time, which would be necessary with escalating doses of dobutamine, was not believed to be warranted. In addition, only a portion of the ventricle was modeled in this study. As a result, this study does not provide information about strain near the base or apex of the heart. Several factors influenced our decision to include only part of the ventricle in our analysis. Currently, short-axis image quality is typically highest in the midventricular region, thus ensuring the highest possible accuracy in our strain measurements. Also, with the approach used in this study, the finite element part of the analysis is completely automated. Efforts are currently under way to allow for automated finite element model construction of the entire ventricle. Additionally, the assessment of only 4 ischemic cardiomyopathic patients limits making significant conclusions regarding the use of this modality for identifying nonviable segments. However, the correlation of abnormal dobutamine angle orientation and nonviability by PET (Table 1) is intriguing and provides a basis for further comparison of vector angle orientation in patients with ischemic cardiomyopathy and the normal data set in this study. Another criticism of cardiac MRI suggested by some investigators is the long scan time and the difficult breath holds for the patients. In our experience, even patients with marked cardiac disease usually tolerate the procedures well. Furthermore, with advances in MRI technology, such as real-time MRI, the scans will become increasingly quicker and hence more adaptable to sicker patients [30, 31].
In this study, the orientation of 3 in the normal left ventricle at rest and during dobutamine administration was characterized by measuring the angle between the minimum principal strain vector and its projection into a plane defined by the local circumferential and longitudinal directions. These angles were small in all regions at rest and were not markedly altered during dobutamine infusion. The consistency of these angle measurements in the normal ventricle, combined with the results of previous research that showed reorientation of the principal strains in dysfunctional myocardium, suggest that these measurements may be useful in accurately regionally characterizing myocardial viability.
Appendix principal strain angle measurement
The orientation of 3 was quantified by measuring the angle between a unit eigenvector associated with 3 at a point and its projection onto a surface that was interpolated from the endocardium and epicardium at that point. Specifically, let 
3 represent the direction of the principal strain 3 at a point P, and consider a surface passing through P obtained by interpolation between the surfaces representing the epicardial and endocardial walls. Let 3 be the projection of 3 into the interpolated surface passing through P (Appendix Fig 1). The principal strain angle, 
is computed from the dot product between 3 and 3:
The projected vector 3' is computed by finding the components of 3 in the local system (uvw) passing through P and setting the v component equal to 0. The u component of the local system runs in the circumferential direction, the v component is transmural, and the w component (longitudinal) is perpendicular to the uv plane.
|
References
|
|---|
- Wu KC, Lima JA. Noninvasive imaging of myocardial viability: current techniques and future developments. Circ Res. 2003;93:1146–1158[Abstract/Free Full Text]
- Shan K, Constantine G, Sivananthan M, Flamm SD. Role of cardiac magnetic resonance imaging in the assessment of myocardial viability. Circulation. 2004;109:1328–1334[Free Full Text]
- Kim RJ, Wu E, Rafael A, et al. The use of contrast-enhanced magnetic resonance imaging to identify reversible myocardial dysfunction. N Engl J Med. 2000;343:1445–1453[Abstract/Free Full Text]
- Gerber BL, Garot J, Bluemke DA, et al. Accuracy of contrast-enhanced magnetic resonance imaging in predicting improvement of regional myocardial function in patients after acute myocardial infarction. Circulation. 2002;106:1083–1089[Abstract/Free Full Text]
- Moustakidis P, Cupps BP, Pomerantz BJ, et al. Noninvasive, quantitative assessment of left ventricular function in ischemic cardiomyopathy. J Surg Res. 2004;116:187–196[Medline]
- Moulton MJ, Creswell LL, Downing SW, et al. Spline surface interpolation for calculating 3-D ventricular strains from MRI tissue tagging. Am J Physiol. 1996;270(1 Pt 2):H281–H297
- Schiller NB, Shah PM, Crawford M, et al. Recommendations for quantitation of the left ventricle by two-dimensional echocardiography. American Society of Echocardiography Committee on Standards, Subcommittee on Quantitation of Two-Dimensional Echocardiograms. J Am Soc Echocardiogr. 1989;2:358–367[Medline]
- Kloner RA, Przyklenk K. Stunned and hibernating myocardium. Annu Rev Med. 1991;42:1–8[Medline]
- Kloner RA, Przyklenk K, Patel B. Altered myocardial states. The stunned and hibernating myocardium. Am J Med. 1989;86:14–22[Medline]
- Young AA, Kramer CM, Ferrari VA, et al. Three-dimensional left ventricular deformation in hypertrophic cardiomyopathy. Circulation. 1994;90:854–867[Abstract/Free Full Text]
- Moore CC, Lugo-Olivieri CH, McVeigh ER, Zerhouni EA. Three-dimensional systolic strain patterns in the normal human left ventricle: characterization with tagged MR imaging. Radiology. 2000;214:453–466[Abstract/Free Full Text]
- Villarreal FJ, Lew WY, Waldman LK, Covell JW. Transmural myocardial deformation in the ischemic canine left ventricle. Circ Res. 1991;68:368–381[Abstract/Free Full Text]
- Marcus JT, Gotte MJ, Van Rossum AC, et al. Myocardial function in infarcted and remote regions early after infarction in man: assessment by magnetic resonance tagging and strain analysis. Magn Reson Med. 1997;38:803–810[Medline]
- Garot J, Bluemke DA, Osman NF, et al. Fast determination of regional myocardial strain fields from tagged cardiac images using harmonic phase MRI. Circulation. 2000;101:981–988[Abstract/Free Full Text]
- Grivegnee AR, Van den Bossche JL, Segebarth G, et al. Magnetic resonance imaging of the heart: a comparison with two dimensional echocardiography. Ann Radiol. 1987;30:105–109
- Baer FM, Voth E, LaRosee K, et al. Comparison of dobutamine transesophageal echocardiography and dobutamine magnetic resonance imaging for detection of residual myocardial viability. Am J Cardiol. 1996;78:415–419[Medline]
- Baer FM, Theissen P, Crnac J, et al. Head to head comparison of dobutamine-transoesophageal echocardiography and dobutamine-magnetic resonance imaging for the prediction of left ventricular functional recovery in patients with chronic coronary artery disease. Eur Heart J. 2000;21:981–991[Abstract/Free Full Text]
- Helbing WA, Bosch HG, Maliepaard C, et al. Comparison of echocardiographic methods with magnetic resonance imaging for assessment of right ventricular function in children. Am J Cardiol. 1995;76:589–594[Medline]
- Camici PG, Wijns W, Borgers M, et al. Pathophysiological mechanisms of chronic reversible left ventricular dysfunction due to coronary artery disease (hibernating myocardium). Circulation. 1997;96:3205–3214[Free Full Text]
- Soufer R, Dey HM, Ng CK, Zaret BL. Comparison of sestamibi single-photon emission computed tomography with positron emission tomography for estimating left ventricular myocardial viability. Am J Cardiol. 1995;75:1214–1219[Medline]
- Maublant JC, Marcaggi X, Lusson JR, et al. Comparison between thallium-201 and technetium-99m methoxyisobutyl isonitrile defect size in single-photon emission computed tomography at rest, exercise and redistribution in coronary artery disease. Am J Cardiol. 1992;69:183–187[Medline]
- Doenst T, Taegtmeyer H. Profound underestimation of glucose uptake by [18F]2-deoxy-2-fluoroglucose in reperfused rat heart muscle. Circulation. 1998;97:2454–2462[Abstract/Free Full Text]
- Axel L, Dougherty L. MR imaging of motion with spatial modulation of magnetization. Radiology. 1989;171:841–845[Abstract/Free Full Text]
- Power TP, Kramer CM, Shaffer AL, et al. Breath-hold dobutamine magnetic resonance myocardial tagging: normal left ventricular response. Am J Cardiol. 1997;80:1203–1207[Medline]
- Young AA, Imai H, Chang CN, Axel L. Two-dimensional left ventricular deformation during systole using magnetic resonance imaging with spatial modulation of magnetization [published erratum appears in Circulation 1994;90:1584]. Circulation. 1994;89:740–752[Abstract/Free Full Text]
- Young AA, Axel L. Three-dimensional motion and deformation of the heart wall: estimation with spatial modulation of magnetization—a model-based approach. Radiology. 1992;185:241–247[Abstract/Free Full Text]
- Zerhouni EA, Parish DM, Rogers WJ, et al. Human heart: tagging with MR imaging—a method for noninvasive assessment of myocardial motion. Radiology. 1988;169:59–63[Abstract/Free Full Text]
- Ungacta FF, Davila-Roman VG, Moulton MJ, et al. MRI-radiofrequency tissue tagging in patients with aortic insufficiency before and after operation. Ann Thorac Surg. 1998;65:943–950[Abstract/Free Full Text]
- Scott CH, Sutton MS, Gusani N, et al. Effect of dobutamine on regional left ventricular function measured by tagged magnetic resonance imaging in normal subjects. Am J Cardiol. 1999;83:412–417[Medline]
- Busch M, Bornstedt A, Wendt M, et al. Fast "real time" imaging with different k-space update strategies for interventional procedures. J Magn Reson Imaging. 1998;8:944–954[Medline]
- Sodickson DK, Manning WJ. Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays. Magn Reson Med. 1997;38:591–603[Medline]
This article has been cited by other articles:
|
|
|
|
 
B. P. Cupps, D. R. Bree, J. R. Wollmuth, A. C. Howells, R. K. Voeller, J. G. Rogers, and M. K. Pasque
Myocardial Viability Mapping by Magnetic Resonance-Based Multiparametric Systolic Strain Analysis
Ann. Thorac. Surg.,
November 1, 2008;
86(5):
1546 - 1553.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
= rest; 
= dobutamine. (Ant = anterior; Lat = lateral; Post = posterior; Sept = septal.)