New Noninvasive Method for Assessment of Left Ventricular Rotation
http://www.100md.com
《循环学杂志》
the Department of Cardiology (T.H.-V., T.E., E.L., H.I., O.A.S.), Rikshospitalet University Hospital
the Department of Circulation and Medical Imaging (J.C., B.H.A., H.T.), Norwegian University of Science and Technology, Trondheim
the Department of Radiology (H.-J.S.), Rikshospitalet University Hospital, Oslo, Norway
the Department of Cardiology (B.D.R., J.A.C.L.), Johns Hopkins University, Baltimore, Md.
Abstract
Background— Left ventricular (LV) torsion is due to oppositely directed apical and basal rotation and has been proposed as a sensitive marker of LV function. In the present study, we introduce and validate speckle tracking echocardiography (STE) as a method for assessment of LV rotation and torsion.
Methods and Results— Apical and basal rotation by STE was measured from short-axis images by automatic frame-to-frame tracking of gray-scale speckle patterns. Rotation was calculated as the average angular displacement of 9 regions relative to the center of a best-fit circle through the same regions. As reference methods we used sonomicrometry in anesthetized dogs during baseline, dobutamine infusion, and apical ischemia, and magnetic resonance imaging (MRI) tagging in healthy humans. In dogs, the mean peak apical rotation was –3.7±1.2° (±SD) and –4.1±1.2°, and basal rotation was 1.9±1.5° and 2.0±1.2° by sonomicrometry and STE, respectively. Rotations by both methods increased (P<0.001) during dobutamine infusion. Apical rotation by both methods decreased during left anterior descending coronary artery occlusion (P<0.007), whereas basal rotation was unchanged. In healthy humans, apical rotation was –11.6±3.8° and –10.9±3.3°, and basal rotation was 4.8±1.7° and 4.6±1.3° by MRI tagging and STE, respectively. Torsion measurement by STE showed good correlation and agreement with sonomicrometry (r=0.94, P<0.001) and MRI (r=0.85, P<0.001).
Conclusions— The present study demonstrates that regional LV rotation and torsion can be measured accurately by STE, suggesting a new echocardiographic approach for quantification of LV systolic function.
Key Words: echocardiography ventricles torsion magnetic resonance imaging rotation
Introduction
Left ventricular (LV) torsion (or twist) plays an important role with respect to LV ejection and filling.1–4 During the cardiac cycle, there is a systolic twist and an early diastolic untwist of the LV about its long axis because of oppositely directed apical and basal rotations. As viewed from the LV apex, systolic apical rotation is counterclockwise and basal rotation, clockwise. The magnitude and characteristics of this torsional deformation have been described in different clinical and experimental studies, and it is well established that LV rotation is sensitive to changes in both regional and global LV function.5–17 Therefore, assessment of LV rotation represents an interesting approach for quantifying LV function. However, so far, magnetic resonance imaging (MRI) tagging has been the only clinically available method,7,13–16,18–21 and implementation has therefore been limited by complexity and cost.
In the present study, we introduce and evaluate echocardiographic speckle tracking (STE) as a new, noninvasive method for assessment of LV rotation and torsion. Because of scattering, reflection, and interference of the ultrasound beams in myocardial tissue, speckle formations in gray-scale echocardiographic images represent tissue markers that can be tracked from frame to frame throughout the cardiac cycle. We hypothesized that STE could be an accurate and clinically applicable noninvasive method for estimating the magnitude and dynamics of LV rotation. In this study, torsion was estimated as the difference in apical and basal rotation, and the STE method was validated by comparison with sonomicrometry in a dog model and with MRI tagging in humans.
Methods
Experimental Study
Thirteen mongrel dogs of either sex and an average body weight of 23.2 kg were anesthetized with a bolus of thiopental 25 mg/kg, followed by continuous infusion of morphine (3.5 mg · kg–1 · h–1) and pentobarbital (2 mg · kg–1 · h–1), the latter reduced to half the dose after 4 hours of infusion. The animals were artificially ventilated through a cuffed endotracheal tube with room air and 20% to 50% O2. The ECG was monitored from limb leads. One femoral vein and 2 jugular arteries were cannulated. After a median sternotomy, the pericardium was split from apex to base and the edges of the pericardial incision were loosely resutured after instrumentation. Inflatable vascular occluders were placed around the proximal third of the left anterior descending coronary artery (LAD). Aortic, left atrial, and LV pressures were measured by micromanometers (MPC-500, Millar Instruments Inc). Data were digitized at 200 Hz. The dogs were placed in the supine position during recordings. The study was approved by the National Animal Experimentation Board. The laboratory animals were supplied by the Center for Comparative Medicine, Rikshospitalet University Hospital, Oslo, Norway.
Sonomicrometry
For estimation of LV torsion and apical and basal rotation, 1 sonomicrometric crystal was implanted at the tip of the apex, and 11 crystals were implanted along the LV circumference at basal (n=3), equatorial (n=4), and apical (n=4) short-axis levels (Figure 1A). Anatomy allowed only 3 crystals at the basal level. To minimize myocardial damage and to achieve reproducible and parallel planes, the crystals at each level were placed subepicardially and at distances 20, 40, and 60 mm from the LV apex. With signals obtained from the 3D grid of crystals, the coordinates of each crystal were automatically determined in space as a function of time (200 Hz). Parallel apical, equatorial, and basal LV planes were constructed by interpolation of the corresponding crystal coordinates, and the in-plane positions were approximated by cubic Hermite interpolation (standard interpolation used in Matlab [The MathWorks Inc]). The center of rotation for each LV plane was determined as the center of a best-fit circle through the interpolated coordinates. For each plane, the angular movements of the interpolated coordinates were averaged, and LV torsion was estimated as the difference in angular movement between apical and basal planes at isochronal points. Apical and basal rotation was calculated by measuring the difference in angular movement between the equatorial level, where rotation is known to be minimal,14,15,19,22–25 and the apical and basal levels, respectively.
Echocardiographic Recordings and Analysis
LV short-axis recordings were obtained by conventional 2D gray-scale echocardiography (Vivid 7 scanner, GE Vingmed). Transducer frequencies (1.7 to 2.0 MHz), sampling rates (70 to 110 frames per second), and sector width (as narrow as possible) were adjusted for optimal speckle quality of the recordings. Short-axis echocardiograms were recorded in the same plane as used for sonomicrometry with the anatomic crystals as a reference. Echocardiographic recordings were done immediately before the sonomicrometric recordings and analyzed with a Matlab-based program that uses the speckle patterns in the gray-scale images. The speckle tracking method, with minimum sums of absolute differences of the B-mode pixel data,26 was used to track the position of a kernel region (a selected region of interest [ROI] with a unique speckle pattern) frame by frame throughout the cardiac cycle. To avoid drift, the tracking algorithm was applied both forward and backward, and the results were averaged. The size of the ROI was 3x5 mm, and the limit for maximum displacement velocity was set to 12 cm/s for vertical and 7 cm/s for lateral velocities. Nine ROIs were automatically superimposed on the echocardiographic image at end diastole and positioned to fit the circle-shaped LV. In our experimental study, this superimposed circle was aligned with the subepicardial LV circumference. Rarely was >1 crystal visible in the echocardiographic image. Therefore, it was not a major problem that nonphysiological speckle patterns would falsely improve tracking quality. Figure 1B shows representative recordings. The starting position of the ROIs could be adjusted manually when the tracking appeared to be poor. ROIs were excluded in regions of insufficient speckle quality because of dropouts of ultrasound data or severe reverberations. ROIs that drifted outside the wall layer being studied were also rejected. In our analyses, 93% of the ROIs remained within the LV layer of interest. Rotation was estimated as the average angular displacement of all 9 ROIs relative to the center of a best-fit circle through the same ROIs, frame by frame. Torsion was estimated as the difference between apical and basal rotation at isochronal points.
Experimental Protocol
After baseline recordings (2 to 3 hours after thoracotomy) were obtained, dobutamine was infused at a rate of 5 μg · kg–1 · min–1, and recordings were repeated. After the return of dP/dt to baseline values, the LAD was occluded for 10 minutes and recordings were obtained. Baseline data were obtained from 13 dogs, whereas dobutamine intervention was not performed in the first 4 experiments. In 5 dogs, recordings during ischemia could not be obtained because of sustained ventricular fibrillation shortly after LAD occlusion. In 6 dogs, short-axis recordings from multiple levels were obtained 5 minutes before and 5 minutes after pericardiotomy. This was done to explore the importance of an intact pericardium for LV rotation.
Clinical Study
Twenty-nine healthy volunteers (15 men and 14 women; mean age, 33±6 years) were included. The study protocol was approved by the National Committee for Medical Research Ethics of Norway. All participants gave written, informed consent.
MRI Tagging
Images were obtained with a 1.5 T scanner (Magnetom Vision Plus, Siemens). To standardize short-axis image planes between individuals, the basal cine image was defined just distal to the fibrous mitral ring, and the apical level, just proximal to the level with luminal closure at end systole. Striped tags were prescribed separately in 2 orthogonal orientations (45° and 135°) with spatial modulation of magnetization in a grid pattern with an 8-mm distance between tags and a time resolution of 35 ms. Images were acquired during 12- to 18-second breath-holds and triggered by ECG. Consistent with STE measurements, rotation by MRI tagging was calculated as the average of measurements obtained in the midendocardial and subendocardial layers. Recordings were analyzed by Harmonic Phase Imaging (HARP version 1.0, Diagnosoft Inc)27 at an experienced MRI reading center (Johns Hopkins University).
Echocardiographic Recordings and Analysis
Short-axis echocardiographic recordings were obtained with the volunteers in a supine left lateral position. The same scanner and acquisition settings were used as for the experimental recordings. Short-axis recordings at the basal level were obtained from a standard parasternal probe position, and recordings at the apical level, from a more distal anterior or anterolateral position. Short-axis images were acquired at approximately the same levels as the MRI cine images. Echocardiographic recordings were taken within 5 to 10 minutes before or after MRI examinations during breath-holds, and an effort was made to make the LV cross section as circular as possible. The quality of the speckles improved progressively from the epicardium to the endocardium, and in many cases, the subepicardial speckle quality was suboptimal. Therefore, we limited the study to assessment of rotation of the midendocardial and subendocardial layers. Criteria for adjustments or rejections of ROIs were the same as for the experimental study.
Reproducibility of the STE Method
All echocardiographic analyses were done without knowledge of the results from the reference methods. To assess interobserver variability, 6 experimental and 6 clinical echocardiographic recordings were randomly selected and then independently analyzed by 2 different observers (including selection of the cardiac cycle, placement of the ROIs, and deriving the results).
Statistical Analysis
Data are presented as mean±SD unless otherwise stated. The rotation and torsion measurements obtained by STE and reference methods were compared by a least-squares linear-regression method and by the Bland-Altman method.28 In the experimental study, we used a 1-way repeated-measures ANOVA followed by the Bonferroni correction for predefined comparisons of baseline versus dobutamine and baseline versus ischemia (SPSS version 12). Statistical differences were considered significant at P<0.05. Interobserver variability was assessed by the intraclass correlation coefficient and the Bland-Altman method.
Results
LV Rotation and Torsion by STE Versus Sonomicrometry: Experimental Study
There was no significant change in heart rate or time to peak rotation from baseline to interventions. However, there was a tendency toward increased heart rate during both ischemia and dobutamine infusion, and a reduced time to peak rotation during dobutamine infusion. Short-axis recordings from the time of crystal implantation confirmed minimal rotation at the equatorial level (–0.2±0.3°).
LV Rotation and Torsion by STE Versus MRI Tagging: Healthy Humans
In our clinical study, ROIs were rejected in only 10% of the recordings (1 ROI in 6% and 2 ROIs in 4% of the recordings). In the experimental study, ROIs were rejected in 18% of the recordings (1 ROI in 10% and 2 ROIs in 8% of the recordings).
Effect of Pericardiotomy on LV Rotation and Torsion
Pericardiotomy caused no change in LV apical or basal rotation (–5.7±0.5 and –5.5±0.4° [P=NS], and 3.1±0.6 and 3.0±1.2° [P=NS], respectively). During the 2- to 3-hour period before the baseline recording, however, apical and basal rotation decreased (to –4.2±0.6, P<0.05, and 1.8±0.7°, P=NS, respectively). Furthermore, LV dP/dtmax and LV pressure, which were unaffected by pericardiotomy (2138±485 and 2134±453 mm Hg/s, P=NS, and 120±7 and 120±6 mm Hg, P=NS, respectively), decreased to 1644±319 (P=NS) and 104±5 (P=0.040), respectively, at the time of baseline recordings. Similarly, torsion was unaffected by pericardiotomy (–8.7±1.8° before versus –8.6±2.0° after) and was reduced at the time of baseline recordings (–6.0±1.5°, P<0.002).
Interobserver Variability
Measurements by STE of peak rotation by 2 independent observers showed a mean difference between the 2 analyses of 0.4±1.6°. The intraclass correlation coefficient between the 2 observers was 0.99. The typical time spent for data analysis was <1 minute.
Discussion
Echocardiography is one of the cornerstones for diagnosing and monitoring cardiac diseases. However, angle-independent assessment of LV deformation and rotation has only been possible by MRI tagging. The present study demonstrates that STE can measure LV torsional deformation noninvasively by automated tracking of speckles from apical and basal short-axis recordings. The validity of our approach was tested with sonomicrometry as a reference method in an animal model and MRI tagging in humans. With sonomicrometry, the implanted myocardial crystals served as anatomic landmarks that ensured that the LV cross-sectional planes studied by the 2 methods were the same, and measurements could be performed only a few seconds apart. Furthermore, in the animal model, comparison between the methods could be completed under a wide range of experimental settings known to alter LV rotation. The STE method showed dynamics, magnitudes, and timing of peak basal and apical rotation and torsion that were closely related to measurements by sonomicrometry. The same relation was found when STE was compared with MRI in healthy volunteers.
Apical Rotation
In the present study, counterclockwise rotation during LV ejection was demonstrated at the apical level in healthy individuals and was found in the dog model. During dobutamine infusion, there was an increase in apical rotation and a tendency toward a decrease in the time to peak rotation, whereas during LAD occlusion, apical rotation was reduced. These changes in apical rotation by dobutamine and apical ischemia are concordant with the findings from previous studies that have used other methods.2,8,10,11,29
Although systolic apical rotation was predominantly counterclockwise, there was a small, clockwise rotation during isovolumic contraction. The oppositely directed rotation was demonstrated by both sonomicrometry and STE but not by MRI (see Figures 3 and 5). This phenomenon has been described previously and might be attributed to earlier activation of subendocardial fibers (right-handed helix) than subepicardial fibers.1,6,30,31 The reason why this motion was not recognized by MRI tagging is probably the relatively low temporal resolution of the method (35 ms).
Basal Rotation
Consistent with previous studies, 14–16,19–21,23–25 LV rotation at the base was predominantly clockwise. During early systole, however, there was a counterclockwise rotation that gradually changed into a more substantial clockwise rotation during ejection. Dobutamine infusion caused a significant increase in basal clockwise rotation, with no change in the time to peak rotation. To our knowledge, no previous studies exist of basal rotation during dobutamine infusion. Apical ischemia caused no significant change in basal rotation,14 reflecting the fact that there was no impairment of LV function between equator and base.
Some studies have indicated that basal LV rotation is minimal.3,12 However, in patients with aortic stenosis, a reduced magnitude of basal rotation compared with normal hearts has been observed.20 Furthermore, in patients with chronic heart failure, 6 months of treatment was associated with an increase in basal rotation, whereas apical rotation was unchanged,16 indicating that measurement of basal rotation may be clinically relevant. However, compared with the reference methods, assessment of rotation by STE was less accurate for the basal level than for the apical level. Therefore, deviations in basal rotation could be difficult to assess by STE, and larger clinical studies will need to be performed to answer this question.
Comparison With Previous Studies
The magnitudes of LV apical and basal rotation and torsion, as reported in previous experimental and clinical studies, differ substantially,2–6,8,9,12–18,20,22–25,29,32–39 and a number of factors may explain this variance. As demonstrated by Henson et al,22 apparent differences in torsion between mice and humans are due to the different sizes of their ventricles. When systolic torsion angle was normalized for LV length, torsion was essentially similar in the 2 species. Furthermore, because torsion angle is a nonlinear function of ventricular length, its magnitude depends critically on the measurement level relative to the LV base or other reference point. The relatively high values of apical rotation measured by Gibbons Kroeker et al3 are most likely explained by their measurement technique, which recorded rotation at the distal part of the LV apex. Another factor that may explain some of the variance in torsion magnitude in different studies is the existence of a marked transmural gradient, with subendocardial values almost twice those of subepicardial values.18 In the dog model, we measured subepicardial rotation, and during baseline conditions, torsion was 6°. These values are in the same range as reported by Buchalter et al,8 who measured both subepicardial and subendocardial torsion in anesthetized dogs by MRI tagging. Furthermore, the hemodynamic and contractile status of the heart in different animal models may vary and give rise to differences in torsion. A significant reduction in rotation at the apical level and a trend toward a reduction at the base was found at the time of baseline recordings. This was associated with reductions in LV pressure and LV dP/dtmax, indicating that there was some reduction in LV systolic function. Most likely the decrease in rotation reflects impairment of LV function owing to the extensive surgical instrumentation.
Limitations
Several important factors may influence the accuracy of STE. The quality of the recordings must be high to achieve correct tracking, and it requires proper adjustment of frame rate, probe frequency, and focus. In the present study, tracking quality was evaluated visually, and ROIs of poor tracking were either manually moved to areas of better speckle quality or deleted. Because rotational displacement is relatively homogenously distributed around the LV circumference in healthy myocardium, deletion of 1 or a maximum of 2 ROIs will have a minimal effect on average rotation.8,18,33
A fundamental problem with STE in LV short-axis images is that longitudinal motion of the LV causes the myocardium to move in and out of the image plane. As a consequence, speckles generated from the ultrasound beam (2 to 3 mm of thickness) will represent the myocardium from different cross-sectional levels during the cardiac cycle. This problem is most pronounced at the LV base, where the longitudinal displacement, according to our results, was 4 mm from end diastole to end systole, whereas the LV apex was essentially stationary. In our study, the relation between STE and the reference methods was best at the apical level, probably because of the aforementioned phenomenon.
A limitation in the clinical study was that speckle quality in some cases was suboptimal in the subepicardial layer of the LV. Therefore, for validation against MRI, we compared measurements from the middle and inner wall layers. Hopefully, technical developments in ultrasound technology will resolve this problem. In the dog study, we had direct access to the heart via sternotomy, and by using ultrasound gel as a standoff, we obtained satisfactory tracking quality in all layers of the LV wall.
One limitation of clinical routine use of STE is the selection of reproducible anatomic landmarks for measuring apical rotation. One approach was to move the cross-sectional image plane as far distally as possible. Another approach could be to measure at the most distal level that does not have luminal closure during systole. At the basal level, reproducible image planes were easier to obtain with the fibrous mitral ring for orientation. Selection of imaging plane is a challenge, and clinical testing of STE in patients is needed to determine whether reproducible measurements can be obtained from ventricles that may change in size and geometry over time.
In contrast to STE, which provides a measure relative to a stationary reference point outside the heart (the echocardiographic transducer), sonomicrometry provides no direct measure of rotation. Because apical and basal rotations are in opposite directions, somewhere between them there exists a level where rotation changes from one direction to the other. A number of studies in dogs, mice, and humans have confirmed that the short-axis level approximately one third from the base to apex (LV equator) shows minimal rotation.14,15,19,22–25 In our study, we measured rotation at the equatorial level by the time of crystal implantation, and the results confirmed minimal rotation (–0.2±2.3°). Taking into account the good relation between STE and sonomicrometry, this finding indicates that our methodological approach was adequate. Furthermore, for strict methodological comparison of STE with sonomicrometry, assessment of LV torsion was sufficient, and the correlation and agreement between torsion by the 2 methods were very good.
Conclusions
This study demonstrates that the magnitude, timing, and dynamics of regional LV rotation and torsion can be measured accurately by STE. When MRI and sonomicrometry were used as reference methods, STE showed good correlation and agreement, suggesting that STE has the potential to become a fast and accurate noninvasive clinical tool.
Acknowledgments
Thomas Helle-Valle and Erik Lyseggen were recipients of clinical research fellowships from the Norwegian Council of Cardiovascular Diseases. Jonas Crosby and Brage H. Amundsen received clinical research fellowships from the Norwegian Research Council. We thank Drs Trond Vartdal, Eirik Pettersen, and Anders Opdahl for beneficial collaboration in the laboratory, and MSc Eldrid Winther-Larssen and engineer Roger Odegaard for technical assistance.
References
McDonald IG. The shape and movements of the human left ventricle during systole: a study by cineangiography and by cineradiography of epicardial markers. Am J Cardiol. 1970; 26: 221–230.
Rademakers FE, Buchalter MB, Rogers WJ, Zerhouni EA, Weisfeldt ML, Weiss JL, Shapiro EP. Dissociation between left ventricular untwisting and filling: accentuation by catecholamines. Circulation. 1992; 85: 1572–1581.
Gibbons Kroeker CA, Ter Keurs HE, Knudtson ML, Tyberg JV, Beyar R. An optical device to measure the dynamics of apex rotation of the left ventricle. Am J Physiol Heart Circ Physiol. 1993; 265: H1444–H1449.
Moon MR, Ingels NB Jr, Daughters GT, Stinson EB, Hansen DE, Miller DC. Alterations in left ventricular twist mechanics with inotropic stimulation and volume loading in human subjects. Circulation. 1994; 89: 142–150.
Hansen DE, Daughters GT, Alderman EL, Ingels NB, Stinson EB, Miller DC. Effect of volume loading, pressure loading, and inotropic stimulation on left ventricular torsion in humans. Circulation. 1991; 83: 1315–1326.
Yun KL, Niczyporuk MA, Daughters GT, Ingels NB Jr, Stinson EB, Alderman EL, Hansen DE, Miller DC. Alterations in left ventricular diastolic twist mechanics during acute human cardiac allograft rejection. Circulation. 1991; 83: 962–973.
Maier SE, Fischer SE, McKinnon GC, Hess OM, Krayenbuehl HP, Boesiger P. Evaluation of left ventricular segmental wall motion in hypertrophic cardiomyopathy with myocardial tagging. Circulation. 1992; 86: 1919–1928.
Buchalter MB, Rademakers FE, Weiss JL, Rogers WJ, Weisfeldt ML, Shapiro EP. Rotational deformation of the canine left ventricle measured by magnetic resonance tagging: effects of catecholamines, ischaemia, and pacing. Cardiovasc Res. 1994; 28: 629–635.
DeAnda A Jr, Komeda M, Nikolic SD, Daughters GT II, Ingels NB, Miller DC. Left ventricular function, twist, and recoil after mitral valve replacement. Circulation. 1995; 92: 458–466.
Gibbons Kroeker CA, Tyberg JV, Beyar R. Effects of load manipulations, heart rate, and contractility on left ventricular apical rotation: an experimental study in anesthetized dogs. Circulation. 1995; 92: 130–141.
Gibbons Kroeker CA, Tyberg JV, Beyar R. Effects of ischemia on left ventricular apex rotation: an experimental study in anesthetized dogs. Circulation. 1995; 92: 3539–3548.
Knudtson ML, Galbraith PD, Hildebrand KL, Tyberg JV, Beyar R. Dynamics of left ventricular apex rotation during angioplasty: a sensitive index of ischemic dysfunction. Circulation. 1997; 96: 801–808.
Stuber M, Scheidegger MB, Fischer SE, Nagel E, Steinemann F, Hess OM, Boesiger P. Alterations in the local myocardial motion pattern in patients suffering from pressure overload due to aortic stenosis. Circulation. 1999; 100: 361–368.
Nagel E, Stuber M, Lakatos M, Scheidegger MB, Boesiger P, Hess OM. Cardiac rotation and relaxation after anterolateral myocardial infarction. Coron Artery Dis. 2000; 11: 261–267.
Sandstede JJW, Johnson T, Harre K, Beer M, Hofmann S, Pabst T, Kenn W, Voelker W, Neubauer S, Hahn D. Cardiac systolic rotation and contraction before and after valve replacement for aortic stenosis: a myocardial tagging study using MR imaging. Am J Roentgenol. 2002; 178: 953–958.
Fuchs E, Muller MF, Oswald H, Thony H, Mohacsi P, Hess OM. Cardiac rotation and relaxation in patients with chronic heart failure. Eur J Heart Failure. 2004; 6: 715–722.
Tibayan FA, Rodriguez F, Langer F, Zasio MK, Bailey L, Liang D, Daughters GT, Ingels NB Jr, Miller DC. Alterations in left ventricular torsion and diastolic recoil after myocardial infarction with and without chronic ischemic mitral regurgitation. Circulation. 2004; 110 (suppl II): II-109–II-114.Abstract.
Buchalter MB, Weiss JL, Rogers WJ, Zerhouni EA, Weisfeldt ML, Beyar R, Shapiro EP. Noninvasive quantification of left ventricular rotational deformation in normal humans using magnetic resonance imaging myocardial tagging. Circulation. 1990; 81: 1236–1244.
Fogel MA, Weinberg PM, Hubbard A, Haselgrove J. Diastolic biomechanics in normal infants utilizing MRI tissue tagging. Circulation. 2000; 102: 218–224.
Nagel E, Stuber M, Burkhard B, Fischer SE, Scheidegger MB, Boesiger P, Hess OM. Cardiac rotation and relaxation in patients with aortic valve stenosis. Eur Heart J. 2000; 21: 582–589.
Setser RM, Kasper JM, Lieber ML, Starling RC, McCarthy PM, White RD. Persistent abnormal left ventricular systolic torsion in dilated cardiomyopathy after partial left ventriculectomy. J Thorac Cardiovasc Surg. 2003; 126: 48–55.
Henson RE, Song SK, Pastorek JS, Ackerman JJH, Lorenz CH. Left ventricular torsion is equal in mice and humans. Am J Physiol Heart Circ Physiol. 2000; 278: H1117–H1123.
Lorenz CH, Pastorek JS, Bundy JM. Delineation of normal human left ventricular twist throughout systole by tagged cine magnetic resonance imaging. J Cardiovasc Magn Reson. 2000; 2: 97–108.
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.
Nichols K, Kamran M, Cooke CD, Faber TL, Garcia EV, Bergmann SR, Depuey EG. Feasibility of detecting cardiac torsion in myocardial perfusion gated SPECT data. J Nucl Cardiol. 2003; 9: 500–507.
Bohs LN, Trahey GE. A novel method for angle independent ultrasonic imaging of blood flow and tissue motion. IEEE Trans Biomed Eng. 1991; 38: 280–286.
Garot J, Bluemke DA, Osman NF, Rochitte CE, McVeigh ER, Zerhouni EA, Prince JL, Lima JAC. Fast determination of regional myocardial strain fields from tagged cardiac images using harmonic phase MRI. Circulation. 2000; 101: 981–988.
Martin Bland J, Altman D. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986; 327: 307–310.
Dong SJ, Hees PS, Huang WM, Buffer SA Jr, Weiss JL, Shapiro EP. Independent effects of preload, afterload, and contractility on left ventricular torsion. Am J Physiol Heart Circ Physiol. 1999; 277: H1053–H1060.
Durrer D, Van Dam RT, Freud GE, Janse MJ, Meijler FL, Arzbaecher RC. Total excitation of isolated human heart. Circulation. 1970; 41: 899–912.
Azhari H, Buchalter M, Sideman S, Shapiro E, Beyar R. A conical model to describe the nonuniformity of the left-ventricular twisting motion. Ann Biomed Eng. 1992; 20: 149–165.
Mirro MJ, Rogers EW, Weyman AE, Feigenbaum H. Angular displacement of the papillary muscles during the cardiac cycle. Circulation. 1979; 60: 327–333.
Hansen DE, Daughters GT, Alderman EL, Ingels NB Jr, Miller DC. Torsional deformation of the left ventricular midwall in human hearts with intramyocardial markers: regional heterogeneity and sensitivity to the inotropic effects of abrupt rate changes. Circ Res. 1988; 62: 941–952.
MacGowan GA, Burkhoff D, Rogers WJ, Salvador D, Azhari H, Hees PS, Zweier JL, Halperin HR, Siu CO, Lima JAC. Effects of afterload on regional left ventricular torsion. Cardiovasc Res. 1996; 31: 917–925.
Rothfeld JM, LeWinter MM, Tischler MD. Left ventricular systolic torsion and early diastolic filling by echocardiography in normal humans. Am J Cardiol. 1998; 81: 1465–1469.
Stuber M, Nagel E, Fischer SE, Spiegel MA, Scheidegger MB, Boesiger P. Quantification of the local heartwall motion by magnetic resonance myocardial tagging. Comput Med Imaging Graph. 1998; 22: 217–228.
Bell SP, Nyland L, Tischler MD, McNabb M, Granzier H, LeWinter MM. Alterations in the determinants of diastolic suction during pacing tachycardia. Circ Res. 2000; 87: 235–240.
Tibayan FA, Yun KL, Fann JI, Lai DT, Timek TA, Daughters GT, Ingels NB, Miller DC. Torsion dynamics in the evolution from acute to chronic mitral regurgitation [erratum appears in J Heart Valve Dis. 2002;11:477]. J Heart Valve Dis. 2002; 11: 39–46.
Sorger JM, Wyman BT, Faris OP, Hunter WC, McVeigh ER. Torsion of the left ventricle during pacing with MRI tagging. J Cardiovasc Magn Reson. 2003; 5: 521–530.(Thomas Helle-Valle, MD; J)
the Department of Circulation and Medical Imaging (J.C., B.H.A., H.T.), Norwegian University of Science and Technology, Trondheim
the Department of Radiology (H.-J.S.), Rikshospitalet University Hospital, Oslo, Norway
the Department of Cardiology (B.D.R., J.A.C.L.), Johns Hopkins University, Baltimore, Md.
Abstract
Background— Left ventricular (LV) torsion is due to oppositely directed apical and basal rotation and has been proposed as a sensitive marker of LV function. In the present study, we introduce and validate speckle tracking echocardiography (STE) as a method for assessment of LV rotation and torsion.
Methods and Results— Apical and basal rotation by STE was measured from short-axis images by automatic frame-to-frame tracking of gray-scale speckle patterns. Rotation was calculated as the average angular displacement of 9 regions relative to the center of a best-fit circle through the same regions. As reference methods we used sonomicrometry in anesthetized dogs during baseline, dobutamine infusion, and apical ischemia, and magnetic resonance imaging (MRI) tagging in healthy humans. In dogs, the mean peak apical rotation was –3.7±1.2° (±SD) and –4.1±1.2°, and basal rotation was 1.9±1.5° and 2.0±1.2° by sonomicrometry and STE, respectively. Rotations by both methods increased (P<0.001) during dobutamine infusion. Apical rotation by both methods decreased during left anterior descending coronary artery occlusion (P<0.007), whereas basal rotation was unchanged. In healthy humans, apical rotation was –11.6±3.8° and –10.9±3.3°, and basal rotation was 4.8±1.7° and 4.6±1.3° by MRI tagging and STE, respectively. Torsion measurement by STE showed good correlation and agreement with sonomicrometry (r=0.94, P<0.001) and MRI (r=0.85, P<0.001).
Conclusions— The present study demonstrates that regional LV rotation and torsion can be measured accurately by STE, suggesting a new echocardiographic approach for quantification of LV systolic function.
Key Words: echocardiography ventricles torsion magnetic resonance imaging rotation
Introduction
Left ventricular (LV) torsion (or twist) plays an important role with respect to LV ejection and filling.1–4 During the cardiac cycle, there is a systolic twist and an early diastolic untwist of the LV about its long axis because of oppositely directed apical and basal rotations. As viewed from the LV apex, systolic apical rotation is counterclockwise and basal rotation, clockwise. The magnitude and characteristics of this torsional deformation have been described in different clinical and experimental studies, and it is well established that LV rotation is sensitive to changes in both regional and global LV function.5–17 Therefore, assessment of LV rotation represents an interesting approach for quantifying LV function. However, so far, magnetic resonance imaging (MRI) tagging has been the only clinically available method,7,13–16,18–21 and implementation has therefore been limited by complexity and cost.
In the present study, we introduce and evaluate echocardiographic speckle tracking (STE) as a new, noninvasive method for assessment of LV rotation and torsion. Because of scattering, reflection, and interference of the ultrasound beams in myocardial tissue, speckle formations in gray-scale echocardiographic images represent tissue markers that can be tracked from frame to frame throughout the cardiac cycle. We hypothesized that STE could be an accurate and clinically applicable noninvasive method for estimating the magnitude and dynamics of LV rotation. In this study, torsion was estimated as the difference in apical and basal rotation, and the STE method was validated by comparison with sonomicrometry in a dog model and with MRI tagging in humans.
Methods
Experimental Study
Thirteen mongrel dogs of either sex and an average body weight of 23.2 kg were anesthetized with a bolus of thiopental 25 mg/kg, followed by continuous infusion of morphine (3.5 mg · kg–1 · h–1) and pentobarbital (2 mg · kg–1 · h–1), the latter reduced to half the dose after 4 hours of infusion. The animals were artificially ventilated through a cuffed endotracheal tube with room air and 20% to 50% O2. The ECG was monitored from limb leads. One femoral vein and 2 jugular arteries were cannulated. After a median sternotomy, the pericardium was split from apex to base and the edges of the pericardial incision were loosely resutured after instrumentation. Inflatable vascular occluders were placed around the proximal third of the left anterior descending coronary artery (LAD). Aortic, left atrial, and LV pressures were measured by micromanometers (MPC-500, Millar Instruments Inc). Data were digitized at 200 Hz. The dogs were placed in the supine position during recordings. The study was approved by the National Animal Experimentation Board. The laboratory animals were supplied by the Center for Comparative Medicine, Rikshospitalet University Hospital, Oslo, Norway.
Sonomicrometry
For estimation of LV torsion and apical and basal rotation, 1 sonomicrometric crystal was implanted at the tip of the apex, and 11 crystals were implanted along the LV circumference at basal (n=3), equatorial (n=4), and apical (n=4) short-axis levels (Figure 1A). Anatomy allowed only 3 crystals at the basal level. To minimize myocardial damage and to achieve reproducible and parallel planes, the crystals at each level were placed subepicardially and at distances 20, 40, and 60 mm from the LV apex. With signals obtained from the 3D grid of crystals, the coordinates of each crystal were automatically determined in space as a function of time (200 Hz). Parallel apical, equatorial, and basal LV planes were constructed by interpolation of the corresponding crystal coordinates, and the in-plane positions were approximated by cubic Hermite interpolation (standard interpolation used in Matlab [The MathWorks Inc]). The center of rotation for each LV plane was determined as the center of a best-fit circle through the interpolated coordinates. For each plane, the angular movements of the interpolated coordinates were averaged, and LV torsion was estimated as the difference in angular movement between apical and basal planes at isochronal points. Apical and basal rotation was calculated by measuring the difference in angular movement between the equatorial level, where rotation is known to be minimal,14,15,19,22–25 and the apical and basal levels, respectively.
Echocardiographic Recordings and Analysis
LV short-axis recordings were obtained by conventional 2D gray-scale echocardiography (Vivid 7 scanner, GE Vingmed). Transducer frequencies (1.7 to 2.0 MHz), sampling rates (70 to 110 frames per second), and sector width (as narrow as possible) were adjusted for optimal speckle quality of the recordings. Short-axis echocardiograms were recorded in the same plane as used for sonomicrometry with the anatomic crystals as a reference. Echocardiographic recordings were done immediately before the sonomicrometric recordings and analyzed with a Matlab-based program that uses the speckle patterns in the gray-scale images. The speckle tracking method, with minimum sums of absolute differences of the B-mode pixel data,26 was used to track the position of a kernel region (a selected region of interest [ROI] with a unique speckle pattern) frame by frame throughout the cardiac cycle. To avoid drift, the tracking algorithm was applied both forward and backward, and the results were averaged. The size of the ROI was 3x5 mm, and the limit for maximum displacement velocity was set to 12 cm/s for vertical and 7 cm/s for lateral velocities. Nine ROIs were automatically superimposed on the echocardiographic image at end diastole and positioned to fit the circle-shaped LV. In our experimental study, this superimposed circle was aligned with the subepicardial LV circumference. Rarely was >1 crystal visible in the echocardiographic image. Therefore, it was not a major problem that nonphysiological speckle patterns would falsely improve tracking quality. Figure 1B shows representative recordings. The starting position of the ROIs could be adjusted manually when the tracking appeared to be poor. ROIs were excluded in regions of insufficient speckle quality because of dropouts of ultrasound data or severe reverberations. ROIs that drifted outside the wall layer being studied were also rejected. In our analyses, 93% of the ROIs remained within the LV layer of interest. Rotation was estimated as the average angular displacement of all 9 ROIs relative to the center of a best-fit circle through the same ROIs, frame by frame. Torsion was estimated as the difference between apical and basal rotation at isochronal points.
Experimental Protocol
After baseline recordings (2 to 3 hours after thoracotomy) were obtained, dobutamine was infused at a rate of 5 μg · kg–1 · min–1, and recordings were repeated. After the return of dP/dt to baseline values, the LAD was occluded for 10 minutes and recordings were obtained. Baseline data were obtained from 13 dogs, whereas dobutamine intervention was not performed in the first 4 experiments. In 5 dogs, recordings during ischemia could not be obtained because of sustained ventricular fibrillation shortly after LAD occlusion. In 6 dogs, short-axis recordings from multiple levels were obtained 5 minutes before and 5 minutes after pericardiotomy. This was done to explore the importance of an intact pericardium for LV rotation.
Clinical Study
Twenty-nine healthy volunteers (15 men and 14 women; mean age, 33±6 years) were included. The study protocol was approved by the National Committee for Medical Research Ethics of Norway. All participants gave written, informed consent.
MRI Tagging
Images were obtained with a 1.5 T scanner (Magnetom Vision Plus, Siemens). To standardize short-axis image planes between individuals, the basal cine image was defined just distal to the fibrous mitral ring, and the apical level, just proximal to the level with luminal closure at end systole. Striped tags were prescribed separately in 2 orthogonal orientations (45° and 135°) with spatial modulation of magnetization in a grid pattern with an 8-mm distance between tags and a time resolution of 35 ms. Images were acquired during 12- to 18-second breath-holds and triggered by ECG. Consistent with STE measurements, rotation by MRI tagging was calculated as the average of measurements obtained in the midendocardial and subendocardial layers. Recordings were analyzed by Harmonic Phase Imaging (HARP version 1.0, Diagnosoft Inc)27 at an experienced MRI reading center (Johns Hopkins University).
Echocardiographic Recordings and Analysis
Short-axis echocardiographic recordings were obtained with the volunteers in a supine left lateral position. The same scanner and acquisition settings were used as for the experimental recordings. Short-axis recordings at the basal level were obtained from a standard parasternal probe position, and recordings at the apical level, from a more distal anterior or anterolateral position. Short-axis images were acquired at approximately the same levels as the MRI cine images. Echocardiographic recordings were taken within 5 to 10 minutes before or after MRI examinations during breath-holds, and an effort was made to make the LV cross section as circular as possible. The quality of the speckles improved progressively from the epicardium to the endocardium, and in many cases, the subepicardial speckle quality was suboptimal. Therefore, we limited the study to assessment of rotation of the midendocardial and subendocardial layers. Criteria for adjustments or rejections of ROIs were the same as for the experimental study.
Reproducibility of the STE Method
All echocardiographic analyses were done without knowledge of the results from the reference methods. To assess interobserver variability, 6 experimental and 6 clinical echocardiographic recordings were randomly selected and then independently analyzed by 2 different observers (including selection of the cardiac cycle, placement of the ROIs, and deriving the results).
Statistical Analysis
Data are presented as mean±SD unless otherwise stated. The rotation and torsion measurements obtained by STE and reference methods were compared by a least-squares linear-regression method and by the Bland-Altman method.28 In the experimental study, we used a 1-way repeated-measures ANOVA followed by the Bonferroni correction for predefined comparisons of baseline versus dobutamine and baseline versus ischemia (SPSS version 12). Statistical differences were considered significant at P<0.05. Interobserver variability was assessed by the intraclass correlation coefficient and the Bland-Altman method.
Results
LV Rotation and Torsion by STE Versus Sonomicrometry: Experimental Study
There was no significant change in heart rate or time to peak rotation from baseline to interventions. However, there was a tendency toward increased heart rate during both ischemia and dobutamine infusion, and a reduced time to peak rotation during dobutamine infusion. Short-axis recordings from the time of crystal implantation confirmed minimal rotation at the equatorial level (–0.2±0.3°).
LV Rotation and Torsion by STE Versus MRI Tagging: Healthy Humans
In our clinical study, ROIs were rejected in only 10% of the recordings (1 ROI in 6% and 2 ROIs in 4% of the recordings). In the experimental study, ROIs were rejected in 18% of the recordings (1 ROI in 10% and 2 ROIs in 8% of the recordings).
Effect of Pericardiotomy on LV Rotation and Torsion
Pericardiotomy caused no change in LV apical or basal rotation (–5.7±0.5 and –5.5±0.4° [P=NS], and 3.1±0.6 and 3.0±1.2° [P=NS], respectively). During the 2- to 3-hour period before the baseline recording, however, apical and basal rotation decreased (to –4.2±0.6, P<0.05, and 1.8±0.7°, P=NS, respectively). Furthermore, LV dP/dtmax and LV pressure, which were unaffected by pericardiotomy (2138±485 and 2134±453 mm Hg/s, P=NS, and 120±7 and 120±6 mm Hg, P=NS, respectively), decreased to 1644±319 (P=NS) and 104±5 (P=0.040), respectively, at the time of baseline recordings. Similarly, torsion was unaffected by pericardiotomy (–8.7±1.8° before versus –8.6±2.0° after) and was reduced at the time of baseline recordings (–6.0±1.5°, P<0.002).
Interobserver Variability
Measurements by STE of peak rotation by 2 independent observers showed a mean difference between the 2 analyses of 0.4±1.6°. The intraclass correlation coefficient between the 2 observers was 0.99. The typical time spent for data analysis was <1 minute.
Discussion
Echocardiography is one of the cornerstones for diagnosing and monitoring cardiac diseases. However, angle-independent assessment of LV deformation and rotation has only been possible by MRI tagging. The present study demonstrates that STE can measure LV torsional deformation noninvasively by automated tracking of speckles from apical and basal short-axis recordings. The validity of our approach was tested with sonomicrometry as a reference method in an animal model and MRI tagging in humans. With sonomicrometry, the implanted myocardial crystals served as anatomic landmarks that ensured that the LV cross-sectional planes studied by the 2 methods were the same, and measurements could be performed only a few seconds apart. Furthermore, in the animal model, comparison between the methods could be completed under a wide range of experimental settings known to alter LV rotation. The STE method showed dynamics, magnitudes, and timing of peak basal and apical rotation and torsion that were closely related to measurements by sonomicrometry. The same relation was found when STE was compared with MRI in healthy volunteers.
Apical Rotation
In the present study, counterclockwise rotation during LV ejection was demonstrated at the apical level in healthy individuals and was found in the dog model. During dobutamine infusion, there was an increase in apical rotation and a tendency toward a decrease in the time to peak rotation, whereas during LAD occlusion, apical rotation was reduced. These changes in apical rotation by dobutamine and apical ischemia are concordant with the findings from previous studies that have used other methods.2,8,10,11,29
Although systolic apical rotation was predominantly counterclockwise, there was a small, clockwise rotation during isovolumic contraction. The oppositely directed rotation was demonstrated by both sonomicrometry and STE but not by MRI (see Figures 3 and 5). This phenomenon has been described previously and might be attributed to earlier activation of subendocardial fibers (right-handed helix) than subepicardial fibers.1,6,30,31 The reason why this motion was not recognized by MRI tagging is probably the relatively low temporal resolution of the method (35 ms).
Basal Rotation
Consistent with previous studies, 14–16,19–21,23–25 LV rotation at the base was predominantly clockwise. During early systole, however, there was a counterclockwise rotation that gradually changed into a more substantial clockwise rotation during ejection. Dobutamine infusion caused a significant increase in basal clockwise rotation, with no change in the time to peak rotation. To our knowledge, no previous studies exist of basal rotation during dobutamine infusion. Apical ischemia caused no significant change in basal rotation,14 reflecting the fact that there was no impairment of LV function between equator and base.
Some studies have indicated that basal LV rotation is minimal.3,12 However, in patients with aortic stenosis, a reduced magnitude of basal rotation compared with normal hearts has been observed.20 Furthermore, in patients with chronic heart failure, 6 months of treatment was associated with an increase in basal rotation, whereas apical rotation was unchanged,16 indicating that measurement of basal rotation may be clinically relevant. However, compared with the reference methods, assessment of rotation by STE was less accurate for the basal level than for the apical level. Therefore, deviations in basal rotation could be difficult to assess by STE, and larger clinical studies will need to be performed to answer this question.
Comparison With Previous Studies
The magnitudes of LV apical and basal rotation and torsion, as reported in previous experimental and clinical studies, differ substantially,2–6,8,9,12–18,20,22–25,29,32–39 and a number of factors may explain this variance. As demonstrated by Henson et al,22 apparent differences in torsion between mice and humans are due to the different sizes of their ventricles. When systolic torsion angle was normalized for LV length, torsion was essentially similar in the 2 species. Furthermore, because torsion angle is a nonlinear function of ventricular length, its magnitude depends critically on the measurement level relative to the LV base or other reference point. The relatively high values of apical rotation measured by Gibbons Kroeker et al3 are most likely explained by their measurement technique, which recorded rotation at the distal part of the LV apex. Another factor that may explain some of the variance in torsion magnitude in different studies is the existence of a marked transmural gradient, with subendocardial values almost twice those of subepicardial values.18 In the dog model, we measured subepicardial rotation, and during baseline conditions, torsion was 6°. These values are in the same range as reported by Buchalter et al,8 who measured both subepicardial and subendocardial torsion in anesthetized dogs by MRI tagging. Furthermore, the hemodynamic and contractile status of the heart in different animal models may vary and give rise to differences in torsion. A significant reduction in rotation at the apical level and a trend toward a reduction at the base was found at the time of baseline recordings. This was associated with reductions in LV pressure and LV dP/dtmax, indicating that there was some reduction in LV systolic function. Most likely the decrease in rotation reflects impairment of LV function owing to the extensive surgical instrumentation.
Limitations
Several important factors may influence the accuracy of STE. The quality of the recordings must be high to achieve correct tracking, and it requires proper adjustment of frame rate, probe frequency, and focus. In the present study, tracking quality was evaluated visually, and ROIs of poor tracking were either manually moved to areas of better speckle quality or deleted. Because rotational displacement is relatively homogenously distributed around the LV circumference in healthy myocardium, deletion of 1 or a maximum of 2 ROIs will have a minimal effect on average rotation.8,18,33
A fundamental problem with STE in LV short-axis images is that longitudinal motion of the LV causes the myocardium to move in and out of the image plane. As a consequence, speckles generated from the ultrasound beam (2 to 3 mm of thickness) will represent the myocardium from different cross-sectional levels during the cardiac cycle. This problem is most pronounced at the LV base, where the longitudinal displacement, according to our results, was 4 mm from end diastole to end systole, whereas the LV apex was essentially stationary. In our study, the relation between STE and the reference methods was best at the apical level, probably because of the aforementioned phenomenon.
A limitation in the clinical study was that speckle quality in some cases was suboptimal in the subepicardial layer of the LV. Therefore, for validation against MRI, we compared measurements from the middle and inner wall layers. Hopefully, technical developments in ultrasound technology will resolve this problem. In the dog study, we had direct access to the heart via sternotomy, and by using ultrasound gel as a standoff, we obtained satisfactory tracking quality in all layers of the LV wall.
One limitation of clinical routine use of STE is the selection of reproducible anatomic landmarks for measuring apical rotation. One approach was to move the cross-sectional image plane as far distally as possible. Another approach could be to measure at the most distal level that does not have luminal closure during systole. At the basal level, reproducible image planes were easier to obtain with the fibrous mitral ring for orientation. Selection of imaging plane is a challenge, and clinical testing of STE in patients is needed to determine whether reproducible measurements can be obtained from ventricles that may change in size and geometry over time.
In contrast to STE, which provides a measure relative to a stationary reference point outside the heart (the echocardiographic transducer), sonomicrometry provides no direct measure of rotation. Because apical and basal rotations are in opposite directions, somewhere between them there exists a level where rotation changes from one direction to the other. A number of studies in dogs, mice, and humans have confirmed that the short-axis level approximately one third from the base to apex (LV equator) shows minimal rotation.14,15,19,22–25 In our study, we measured rotation at the equatorial level by the time of crystal implantation, and the results confirmed minimal rotation (–0.2±2.3°). Taking into account the good relation between STE and sonomicrometry, this finding indicates that our methodological approach was adequate. Furthermore, for strict methodological comparison of STE with sonomicrometry, assessment of LV torsion was sufficient, and the correlation and agreement between torsion by the 2 methods were very good.
Conclusions
This study demonstrates that the magnitude, timing, and dynamics of regional LV rotation and torsion can be measured accurately by STE. When MRI and sonomicrometry were used as reference methods, STE showed good correlation and agreement, suggesting that STE has the potential to become a fast and accurate noninvasive clinical tool.
Acknowledgments
Thomas Helle-Valle and Erik Lyseggen were recipients of clinical research fellowships from the Norwegian Council of Cardiovascular Diseases. Jonas Crosby and Brage H. Amundsen received clinical research fellowships from the Norwegian Research Council. We thank Drs Trond Vartdal, Eirik Pettersen, and Anders Opdahl for beneficial collaboration in the laboratory, and MSc Eldrid Winther-Larssen and engineer Roger Odegaard for technical assistance.
References
McDonald IG. The shape and movements of the human left ventricle during systole: a study by cineangiography and by cineradiography of epicardial markers. Am J Cardiol. 1970; 26: 221–230.
Rademakers FE, Buchalter MB, Rogers WJ, Zerhouni EA, Weisfeldt ML, Weiss JL, Shapiro EP. Dissociation between left ventricular untwisting and filling: accentuation by catecholamines. Circulation. 1992; 85: 1572–1581.
Gibbons Kroeker CA, Ter Keurs HE, Knudtson ML, Tyberg JV, Beyar R. An optical device to measure the dynamics of apex rotation of the left ventricle. Am J Physiol Heart Circ Physiol. 1993; 265: H1444–H1449.
Moon MR, Ingels NB Jr, Daughters GT, Stinson EB, Hansen DE, Miller DC. Alterations in left ventricular twist mechanics with inotropic stimulation and volume loading in human subjects. Circulation. 1994; 89: 142–150.
Hansen DE, Daughters GT, Alderman EL, Ingels NB, Stinson EB, Miller DC. Effect of volume loading, pressure loading, and inotropic stimulation on left ventricular torsion in humans. Circulation. 1991; 83: 1315–1326.
Yun KL, Niczyporuk MA, Daughters GT, Ingels NB Jr, Stinson EB, Alderman EL, Hansen DE, Miller DC. Alterations in left ventricular diastolic twist mechanics during acute human cardiac allograft rejection. Circulation. 1991; 83: 962–973.
Maier SE, Fischer SE, McKinnon GC, Hess OM, Krayenbuehl HP, Boesiger P. Evaluation of left ventricular segmental wall motion in hypertrophic cardiomyopathy with myocardial tagging. Circulation. 1992; 86: 1919–1928.
Buchalter MB, Rademakers FE, Weiss JL, Rogers WJ, Weisfeldt ML, Shapiro EP. Rotational deformation of the canine left ventricle measured by magnetic resonance tagging: effects of catecholamines, ischaemia, and pacing. Cardiovasc Res. 1994; 28: 629–635.
DeAnda A Jr, Komeda M, Nikolic SD, Daughters GT II, Ingels NB, Miller DC. Left ventricular function, twist, and recoil after mitral valve replacement. Circulation. 1995; 92: 458–466.
Gibbons Kroeker CA, Tyberg JV, Beyar R. Effects of load manipulations, heart rate, and contractility on left ventricular apical rotation: an experimental study in anesthetized dogs. Circulation. 1995; 92: 130–141.
Gibbons Kroeker CA, Tyberg JV, Beyar R. Effects of ischemia on left ventricular apex rotation: an experimental study in anesthetized dogs. Circulation. 1995; 92: 3539–3548.
Knudtson ML, Galbraith PD, Hildebrand KL, Tyberg JV, Beyar R. Dynamics of left ventricular apex rotation during angioplasty: a sensitive index of ischemic dysfunction. Circulation. 1997; 96: 801–808.
Stuber M, Scheidegger MB, Fischer SE, Nagel E, Steinemann F, Hess OM, Boesiger P. Alterations in the local myocardial motion pattern in patients suffering from pressure overload due to aortic stenosis. Circulation. 1999; 100: 361–368.
Nagel E, Stuber M, Lakatos M, Scheidegger MB, Boesiger P, Hess OM. Cardiac rotation and relaxation after anterolateral myocardial infarction. Coron Artery Dis. 2000; 11: 261–267.
Sandstede JJW, Johnson T, Harre K, Beer M, Hofmann S, Pabst T, Kenn W, Voelker W, Neubauer S, Hahn D. Cardiac systolic rotation and contraction before and after valve replacement for aortic stenosis: a myocardial tagging study using MR imaging. Am J Roentgenol. 2002; 178: 953–958.
Fuchs E, Muller MF, Oswald H, Thony H, Mohacsi P, Hess OM. Cardiac rotation and relaxation in patients with chronic heart failure. Eur J Heart Failure. 2004; 6: 715–722.
Tibayan FA, Rodriguez F, Langer F, Zasio MK, Bailey L, Liang D, Daughters GT, Ingels NB Jr, Miller DC. Alterations in left ventricular torsion and diastolic recoil after myocardial infarction with and without chronic ischemic mitral regurgitation. Circulation. 2004; 110 (suppl II): II-109–II-114.Abstract.
Buchalter MB, Weiss JL, Rogers WJ, Zerhouni EA, Weisfeldt ML, Beyar R, Shapiro EP. Noninvasive quantification of left ventricular rotational deformation in normal humans using magnetic resonance imaging myocardial tagging. Circulation. 1990; 81: 1236–1244.
Fogel MA, Weinberg PM, Hubbard A, Haselgrove J. Diastolic biomechanics in normal infants utilizing MRI tissue tagging. Circulation. 2000; 102: 218–224.
Nagel E, Stuber M, Burkhard B, Fischer SE, Scheidegger MB, Boesiger P, Hess OM. Cardiac rotation and relaxation in patients with aortic valve stenosis. Eur Heart J. 2000; 21: 582–589.
Setser RM, Kasper JM, Lieber ML, Starling RC, McCarthy PM, White RD. Persistent abnormal left ventricular systolic torsion in dilated cardiomyopathy after partial left ventriculectomy. J Thorac Cardiovasc Surg. 2003; 126: 48–55.
Henson RE, Song SK, Pastorek JS, Ackerman JJH, Lorenz CH. Left ventricular torsion is equal in mice and humans. Am J Physiol Heart Circ Physiol. 2000; 278: H1117–H1123.
Lorenz CH, Pastorek JS, Bundy JM. Delineation of normal human left ventricular twist throughout systole by tagged cine magnetic resonance imaging. J Cardiovasc Magn Reson. 2000; 2: 97–108.
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.
Nichols K, Kamran M, Cooke CD, Faber TL, Garcia EV, Bergmann SR, Depuey EG. Feasibility of detecting cardiac torsion in myocardial perfusion gated SPECT data. J Nucl Cardiol. 2003; 9: 500–507.
Bohs LN, Trahey GE. A novel method for angle independent ultrasonic imaging of blood flow and tissue motion. IEEE Trans Biomed Eng. 1991; 38: 280–286.
Garot J, Bluemke DA, Osman NF, Rochitte CE, McVeigh ER, Zerhouni EA, Prince JL, Lima JAC. Fast determination of regional myocardial strain fields from tagged cardiac images using harmonic phase MRI. Circulation. 2000; 101: 981–988.
Martin Bland J, Altman D. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986; 327: 307–310.
Dong SJ, Hees PS, Huang WM, Buffer SA Jr, Weiss JL, Shapiro EP. Independent effects of preload, afterload, and contractility on left ventricular torsion. Am J Physiol Heart Circ Physiol. 1999; 277: H1053–H1060.
Durrer D, Van Dam RT, Freud GE, Janse MJ, Meijler FL, Arzbaecher RC. Total excitation of isolated human heart. Circulation. 1970; 41: 899–912.
Azhari H, Buchalter M, Sideman S, Shapiro E, Beyar R. A conical model to describe the nonuniformity of the left-ventricular twisting motion. Ann Biomed Eng. 1992; 20: 149–165.
Mirro MJ, Rogers EW, Weyman AE, Feigenbaum H. Angular displacement of the papillary muscles during the cardiac cycle. Circulation. 1979; 60: 327–333.
Hansen DE, Daughters GT, Alderman EL, Ingels NB Jr, Miller DC. Torsional deformation of the left ventricular midwall in human hearts with intramyocardial markers: regional heterogeneity and sensitivity to the inotropic effects of abrupt rate changes. Circ Res. 1988; 62: 941–952.
MacGowan GA, Burkhoff D, Rogers WJ, Salvador D, Azhari H, Hees PS, Zweier JL, Halperin HR, Siu CO, Lima JAC. Effects of afterload on regional left ventricular torsion. Cardiovasc Res. 1996; 31: 917–925.
Rothfeld JM, LeWinter MM, Tischler MD. Left ventricular systolic torsion and early diastolic filling by echocardiography in normal humans. Am J Cardiol. 1998; 81: 1465–1469.
Stuber M, Nagel E, Fischer SE, Spiegel MA, Scheidegger MB, Boesiger P. Quantification of the local heartwall motion by magnetic resonance myocardial tagging. Comput Med Imaging Graph. 1998; 22: 217–228.
Bell SP, Nyland L, Tischler MD, McNabb M, Granzier H, LeWinter MM. Alterations in the determinants of diastolic suction during pacing tachycardia. Circ Res. 2000; 87: 235–240.
Tibayan FA, Yun KL, Fann JI, Lai DT, Timek TA, Daughters GT, Ingels NB, Miller DC. Torsion dynamics in the evolution from acute to chronic mitral regurgitation [erratum appears in J Heart Valve Dis. 2002;11:477]. J Heart Valve Dis. 2002; 11: 39–46.
Sorger JM, Wyman BT, Faris OP, Hunter WC, McVeigh ER. Torsion of the left ventricle during pacing with MRI tagging. J Cardiovasc Magn Reson. 2003; 5: 521–530.(Thomas Helle-Valle, MD; J)