- Email: [email protected]
Left Ventricular Twist Mechanics in Hypertrophic Cardiomyopathy Assessed by Three-Dimensional Speckle Tracking Echocardiography
Left Ventricular Twist Mechanics in Hypertrophic Cardiomyopathy Assessed by Three-Dimensional Speckle Tracking Echocardiography Jose A. Urbano Moral, MD*, Jose A. Arias Godinez, MD, Martin S. Maron, MD, Rabiya Malik, MD, Jacqueline E. Eagan, BS, Ayan R. Patel, MD, and Natesa G. Pandian, MD Left ventricular (LV) twist represents a phenomenon that links systolic contraction with diastolic relaxation and plays a major role in cardiac physiology; thus, the study of twist mechanics is of particular interest in hypertrophic cardiomyopathy (HC). Three-dimensional speckle tracking echocardiography (3D-STE) has the potential to overcome the limitations of 2-dimensional imaging and provide a greater understanding of LV twist in HC. We aimed to examine LV twist mechanics in HC using 3D-STE. Echocardiograms from subjects with a diagnosis of HC were examined for 3D-STE analysis. Age- and gender-matched healthy subjects were tested as a control group. Forty patients with HC (age 37 ⴞ 16 years; 42.5% women) and 40 control subjects (age 35 ⴞ 10 years; 42.5% women) were examined. Compared with the controls, the patients with HC showed increased peak LV twist (16.5 ⴞ 4.7° vs 12.0 ⴞ 3.9°, p <0.001) mainly because of increased apical rotation of those with LV outflow tract obstruction (obstruction, 12.7 ⴞ 4.4° vs nonobstruction, 9.7 ⴞ 2.8°, p ⴝ 0.02). In addition, the patients with HC displayed onset of torsion recoil occurring closer to the aortic valve closure (94 ⴞ 6% vs 85 ⴞ 6%, p <0.001; time normalized by the length of systole), limited completion of untwist during early diastole (31 ⴞ 12% vs 62 ⴞ 15%, p <0.001), and delayed peak untwist velocity (22 ⴞ 7% vs 13 ⴞ 9%, p <0.001; time normalized by the length of diastole). In conclusion, the evaluation of twist mechanics using 3D-STE provides novel insight regarding alterations in LV mechanics in patients with HC. Elucidating the characteristics of the wringing motion of the heart might help to broaden the understanding of the hyperdynamic contraction and impaired relaxation observed in these patients. © 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;108:1788 –1795) Left ventricular (LV) twist represents a phenomenon that links systolic contraction with diastolic relaxation and plays a major role in cardiac physiology.1– 4 Speckle tracking echocardiography (STE), based on tracking and measurement of tissue displacement, has the potential for accurate and reliable assessment of myocardial mechanics,5 providing a relatively simple, noninvasive approach to the study of LV rotation and twist.6 The presence, in patients with hypertrophic cardiomyopathy (HC), of a supranormal ejection fraction within a clearly diseased myocardium, and important limitations of transmitral velocities and tissue Doppler imaging for the estimation of filling abnormalities7 has generated interest in evaluating the usefulness of STE for HC,8 –10 where its application might help to widen the understanding of the complex pathophysiology of HC. Despite
From the Cardiovascular Imaging and Hemodynamic Laboratory, Tufts Medical Center, Boston Massachusetts. Manuscript received May 30, 2011; manuscript received and accepted July 20, 2011. The Cardiovascular Imaging and Hemodynamic Laboratory at Tufts Medical Center (Boston, Massachusetts) received an equipment grant from Toshiba Medical Systems (Tustin, California). *Corresponding author: Tel: 617-636-6151; fax: 617-636-8070. E-mail address: [email protected] (J. A. Urbano Moral). 0002-9149/11/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2011.07.047
interest in STE for the assessment of HC through rotational and twist parameters, the available data are quite scarce and limited to a few reports,8 –11 all of which used 2-dimensional technology, with the inherent limitation of tracking out-ofplane tissue motion.12 The aim of the present study was to describe LV twist mechanics in HC using 3-dimensional (3D)-STE. Methods We examined outpatients referred to our echocardiography laboratory from the Hypertrophic Cardiomyopathy Center (Tufts Medical Center, Boston, Massachusetts). Inclusion required a diagnosis of HC (the demonstration of a hypertrophied nondilated left ventricle [in adults, a wall thickness of ⱖ15 mm; in children, a wall thickness ⬎⫹2 SDs of normal values according to body surface area13]) and the absence of hypertension, diabetes mellitus, or any other cardiac or systemic disease that could produce the magnitude of hypertrophy evident. In addition, sinus rhythm had to be present at echocardiography. Left ventricular outflow tract obstruction was considered when an at rest or provocable (Valsalva maneuver) gradient of ⱖ30 mm Hg was detected. The patients were questioned in detail about their history and symptoms of coronary artery disease and were excluded if they provided any history suggesting this condition www.ajconline.org
Cardiomyopathy/Twist Mechanics in Hypertrophic Cardiomyopathy
or if evidence of coronary artery disease was identified by a review of the medical records. Patients demonstrating wall motion abnormalities on the standard echocardiogram were also excluded. In addition, as a part of the routine assessment of HC in our institution, 18 of 40 enrolled patients (those without at rest or Valsalva-induced obstruction) underwent a treadmill echocardiographic test to evaluate for exercise-provocable obstruction. All these patients reached the target heart rate, with neither ST-segment changes nor wall motion abnormalities on the peak exercise echocardiogram. Finally, patients with previous septal myectomy or alcohol septal ablation were excluded. Normal age- and gender-matched volunteers were examined as controls (⫾5 years matching for age and exact matching for gender). They were selected as normal subjects after a detailed history was obtained regarding possible health issues, with an absence of complaints compatible with cardiac disease and no history of hypertension, diabetes mellitus, or any other systemic disease. They were excluded as controls if any of the following was detected on the standard echocardiogram: significant valvulopathy; abnormal cardiac chamber size; or ejection fraction ⬍50%. The institutional review board approved the present study. Echocardiograms were performed with the Artida 4D System scanner (Toshiba Medical Systems, Tustin, California) in the tissue harmonic mode. The PST-30SBT transducer was used for standard 2D/Doppler measurements and the matrix array PST-25SX transducer for acquisition of 3D data sets. The latter consisted of an apical full volume created by the combination of 6 wedge-shaped subvolumes from 6 consecutive cardiac cycles during a single breathhold; 2 to 4 data sets with a temporal resolution of 20 to 25 volumes per second were obtained from each patient to subsequently select that of the best quality for off-line processing. The full-volume 3D data sets were analyzed using the 3D Wall Motion Tracking software (Toshiba Medical Systems) by an expert in the interpretation of echocardiographic images. First, the apical 4- and 2-chamber views and the 3 short axis views were automatically selected at end-diastole. Then, the foreshortened images were avoided by looking for the largest long axis dimensions. The transversal axes were modified as appropriate to optimize their orientation throughout the cardiac cycle. Next, the LV endocardial surface at end-diastole was automatically traced; only in the cases in which the software algorithm did not identify the surface correctly, manual tracing was used. Finally, myocardial tissue displacement was tracked throughout the cardiac cycle. The left ventricle was divided into 16 segments.14 The curves and values of LV rotation and twist and the timing of these parameters were provided by the software. Untwist and untwist rate were calculated by the investigators using the following formulas: Untwist ⫽ 关共 PLVT ⫺ Twist1兲 ⁄ PLVT兴 ⫻ 100 Untwist rate ⫽ 共Twist1 ⫺ PLVT兲 ⁄
关共t1 ⫺ tPLVT兲
⁄ 1000兴
where PLVT is the peak LV twist, Twist1 represents the degree of twist at a specific point along diastole, and t1 and tPLVT stand for the elapsed times at a specific point along
1789
Table 1 Baseline characteristics in patients with hypertrophic cardiomyopathy (HC) and controls Parameter Age (years) Women Heart rate (beats/min) Maximum left ventricular wall thickness (mm) Pattern of hypertrophy Septal asymmetric Symmetric (concentric) Predominantly midcavity New York Heart Association functional class Mitral regurgitation Number/trace Mild Moderate Severe Presence of left ventricular outflow tract obstruction Left ventricular outflow tract gradient at rest (mm Hg) History of paroxysmal atrial fibrillation Medications  Blockers Calcium channel blockers Disopyramide Diuretics
Controls (n ⫽ 40)
HC (n ⫽ 40)
p Value
35 ⫾ 10 17 (42.5%) 67 ⫾ 10 8⫾2
37 ⫾ 16 17 (42.5%) 66 ⫾ 12 21 ⫾ 4
0.5 1.0 0.7 ⬍0.001
35 (87.5%) 2 (5%) 3 (7.5%) 1.8 ⫾ 0.8
21 (52.5%) 12 (30%) 5 (12.5%) 2 (5%) 22 (55%) 42 ⫾ 38 7 (17.5%)
28 (70%) 27 (67.5%) 1 (2.5%) 1 (2.5%)
Continuous data are presented as mean ⫾ SD and categorical data are presented as n (%).
diastole and at PLVT, respectively. In our study, the untwist rate was calculated for mitral valve opening (difference in the degree of twist between mitral valve opening and peak LV twist divided into the corresponding interval). The denominator was divided by 1,000 to express the units in degrees/second. The instantaneous twist and untwist velocities were calculated through the difference in the degree of twist and time ratio between pairs of consecutive frames: Twist/untwist velocity ⫽ 共Twist1 ⫺ Twist0兲 ⁄
关共t1 ⫺ t0兲
⁄ 1000兴
The data are presented as the mean ⫾ SD or frequencies (percentages). An unpaired Student’s t test was used for comparisons between the normal subjects and those with HC, and between HC patients with and without outflow tract obstruction. The ability of the determined parameters to discriminate the patients with HC from the controls was evaluated by receiver operating characteristic curves. The optimal cutoff values were defined as those points providing the greatest sum of sensitivity and specificity. The role of the PLVT for predicting the untwist rate was assessed using simple linear regression analysis. The correlations were assessed using Pearson’s correlation coefficient, and the comparison between them by Fisher’s Z transformation. Statistical significance was defined by a 2-tailed p value of
1790
The American Journal of Cardiology (www.ajconline.org)
Table 2 Parameters derived from 3-dimensional speckle tracking analysis (3D-STE) Parameter End-diastolic volume (ml/m2) End-systolic volume (ml/m2) Ejection fraction (%) Left ventricular mass (g/m2) Peak apical rotation (°) Time to peak apical rotation§ (%) Peak basal rotation (°) Time to peak basal rotation§ (%) Peak left ventricular twist (°) Time to peak left ventricular twist§ (%) Peak twist velocity (°/s) Time to peak twist velocity§ (%)
Controls (n ⫽ 40)
All HC (n ⫽ 40)
p Value
Obstructive* HC (n ⫽ 22)
Nonobstructive* HC (n ⫽ 18)
p Value
62 ⫾ 9 25 ⫾ 9 59 ⫾ 3 73 ⫾ 11 8.2 ⫾ 3.6 83 ⫾ 8 ⫺2.6 ⫾ 1.6 87 ⫾ 7 12.0 ⫾ 3.9 85 ⫾ 6 62 ⫾ 19 57 ⫾ 13
62 ⫾ 13 23 ⫾ 5 63 ⫾ 4 104 ⫾ 26 11.3 ⫾ 4.0 91 ⫾ 8 ⫺3.7 ⫾ 1.9 92 ⫾ 10 16.5 ⫾ 4.7 94 ⫾ 6 77 ⫾ 19 57 ⫾ 19
0.9 0.03 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 0.005 0.007 ⬍0.001 ⬍0.001 0.001 0.9
63 ⫾ 13 23 ⫾ 6 63 ⫾ 4 109 ⫾ 28 12.7 ⫾ 4.4† 95 ⫾ 6 ⫺3.5 ⫾ 2.1¶ 90 ⫾ 10 18.4 ⫾ 4.6† 93 ⫾ 6 81 ⫾ 19 58 ⫾ 20
62 ⫾ 12 22 ⫾ 5 63 ⫾ 4 99 ⫾ 23 9.7 ⫾ 2.8‡ 88 ⫾ 10 ⫺3.9 ⫾ 1.5储 95 ⫾ 9 14.1 ⫾ 3.6# 94 ⫾ 7 72 ⫾ 18 56 ⫾ 17
0.8 0.7 0.9 0.3 0.02 0.01 0.6 0.08 0.003 0.9 0.2 0.8
Data are presented as mean ⫾ SD. * Left ventricular outflow tract obstruction and no left ventricular outflow tract obstruction. † p ⬍0.001 versus controls. ‡ p ⫽ 0.1 versus controls. ¶ p ⫽ 0.049 versus controls. 储 p ⫽ 0.005 versus controls. # p ⫽ 0.049 versus controls. § Time normalized by the length of systole.
⬍0.05. The inter- and intraobserver variability, tested by intraclass correlation coefficient, was as follows: PLVT 0.85 (95% confidence interval [CI] 0.69 to 0.94) and 0.89 (95% CI 0.71 to 0.97); untwist rate 0.88 (95% CI 0.72 to 0.96) and 0.90 (95% CI 0.72 to 0.97); and untwist at mitral valve opening 0.89 (95% CI 0.74 to 0.97) and 0.91 (95% CI 0.75 to 0.98). Results A total of 48 subjects with a diagnosis of HC were initially examined for eligibility for the present study. Inadequate visualization of the endocardial boundary precluded correct wall motion tracking in 8, who were eventually excluded from the analysis. The remaining 40 showed adequate tracking in all 16 LV segments and were consecutively included (Table 1). Up to 50 healthy volunteers (10 without appropriate wall motion tracking) were examined to form the control group. The 3D-STE parameters are summarized in Table 2. A systolic wringing motion with clockwise basal and counterclockwise apical rotation occurred in both groups. Peak LV twist showed significantly greater values in the HC group (16.5 ⫾ 4.7° vs 12.0 ⫾ 3.9°, p ⬍0.001). Figure 1 demonstrates examples of rotational and twist curves from the control and HC groups. The comparisons between HC patients with and without outflow tract obstruction showed differences in apical rotation and PLVT (Tables 2 and 3). In patients with obstruction, no significant correlations were found between those parameters and the at rest LV outflow tract gradient (apical rotation, r ⫽ 0.17, p ⫽ 0.4; PLVT, r ⫽ ⫺0.05, p ⫽ 0.8). Peak LV twist, and therefore, onset of LV untwist occurred later in those with HC (Table 2). From that point, untwist rate was slower in the HC group (Table 3), which was associated with only 31 ⫾ 12% of untwist at mitral
valve opening versus 62 ⫾ 15% in the control subjects (p ⬍0.001). Furthermore, the patients with HC expended 67 ⫾ 23% of diastole to reach 90% of untwist, an event that occurred at 45 ⫾ 21% of diastole in the controls (p ⬍0.001). Different parameter distributions and optimal cutoff values to discriminate those with HC from the controls are displayed in Figures 2 and 3. Although no significant difference was observed for peak untwist velocity (PUV), it occurred later during diastole in the HC group (Table 3 and Figure 4). Moreover, although the timing of PUV and mitral valve opening did not differ in the control group (13 ⫾ 9% vs 15 ⫾ 3%, p ⫽ 0.2; time normalized by the length of diastole), in the patients with HC, the former was significantly delayed with respect to isovolumic relaxation period (22 ⫾ 7% vs 17 ⫾ 4%, p ⬍0.001; time normalized by the length of diastole). The simple linear regression analysis found PLVT to be associated with the untwist rate at mitral valve opening in both controls and those with HC, with a moderate correlation between those parameters (Figure 5). No statistically significant difference was found between the respective correlation coefficients associated with the HC and control groups (p ⫽ 0.7). Discussion LV twist has a relevant role in maintaining efficient myocardial contraction and aids in generating suction power in early diastole.1– 4 The present study, using 3D-STE, allowed us to state the following observations: patients with HC showed increased PLVT that was mainly dependent on the apical rotation of those with LV outflow tract obstruction, onset of torsion recoil occurring close to the aortic valve closure, limited completion of untwist during early diastole, and delayed PUV. Initially, STE analysis of the patients with HC showed a mild rotation gradient between the LV apex and base.8
Cardiomyopathy/Twist Mechanics in Hypertrophic Cardiomyopathy
1791
Figure 1. Rotation (Top) and twist (Bottom) curves in control subject (A,C) and patient with HC (B,D). Apical and basal levels of patient with HC (B) showed greater peak rotation than those seen in control subject (A), evident in color-coded LV chamber schematics attached to each figure. In contrast, patient with HC featured greater PLVT (D), close to aortic valve closure. This was followed by lower degree of untwist at mitral valve opening compared to nearly complete untwist in control subject (C). AVC ⫽ aortic valve closure; MVO ⫽ mitral valve opening.
Subsequent reports suggested that this difference might be greater and especially influenced by the pattern of hypertrophy.9,11 In our HC group, most patients showed septal asymmetric hypertrophy involving primarily the mid and basal levels, points that should be considered when interpreting the prominent apical-basal rotation gradient found in these patients. In contrast, a greater apical and basal rotation in HC compared with healthy
controls has been reported by others.9 Our results, in keeping with those observations, suggest that although the basal ventricle is a major location for extreme hypertrophy, it maintains its contribution to the wringing motion during systole. In this regard, enhanced contractility has been demonstrated to markedly increase torsion15–17; therefore, the hypercontractile state in HC might explain the extended twist mechanics in these patients. Further-
1792
The American Journal of Cardiology (www.ajconline.org)
Table 3 Comparisons of untwist parameters in obstructive and nonobstructive hypertrophic cardiomyopathy (HC) and control groups Parameter
Controls (n ⫽ 40)
All HC (n ⫽ 40)
p Value
Obstructive* HC (n ⫽ 22)
Nonobstructive* HC (n ⫽ 18)
p Value
Isovolumic relaxation time (ms) Untwist at mitral valve opening (%) Untwist rate at mitral valve opening (°/s) Time to 90% of untwist† (%) Peak untwist velocity (°/s) Time to peak untwist velocity† (%) Time to mitral valve opening† (%)
79 ⫾ 8 62 ⫾ 15 ⫺56 ⫾ 23 45 ⫾ 21 ⫺78 ⫾ 27 13 ⫾ 9‡ 15 ⫾ 3
93 ⫾ 12 31 ⫾ 12 ⫺44 ⫾ 20 67 ⫾ 23 ⫺84 ⫾ 28 22 ⫾ 7§ 17 ⫾ 4
⬍0.001 ⬍0.001 0.01 ⬍0.001 0.3 ⬍0.001 0.008
95 ⫾ 12 31 ⫾ 12 ⫺48 ⫾ 21 72 ⫾ 21 ⫺90 ⫾ 31 22 ⫾ 7 17 ⫾ 3
91 ⫾ 13 31 ⫾ 12 ⫺39 ⫾ 17 60 ⫾ 25 ⫺77 ⫾ 24 23 ⫾ 7 17 ⫾ 5
0.4 0.9 0.2 0.1 0.2 0.7 0.9
Data are presented as mean ⫾ SD. * Left ventricular outflow tract obstruction and no left ventricular outflow tract obstruction. † Time normalized by the length of diastole. ‡ p ⫽ 0.2 versus time to mitral valve opening. § p ⬍0.001 versus time to mitral valve opening.
Figure 2. Distribution in peak LV twist, untwist rate at mitral valve opening, untwist at mitral valve opening, and peak untwist velocity related to controls and those with HC. Box plots defined by median (line in center of box), lower and upper quartiles (Q1 and Q3, holding 75% of central values in studied variable), minimum and maximum value within 1.5 interquartile range of lower and upper quartile, respectively (caps at end of each box), and values out of these limits (outliers, small circles).
Cardiomyopathy/Twist Mechanics in Hypertrophic Cardiomyopathy
1793
Figure 3. Receiver operating characteristic curves for those parameters with significant intergroup differences in Figure 2. Cutoff values that yielded greatest sensitivity and specificity for each parameter were as follows: PLVT of 13.3°, sensitivity 75% and specificity 73%; untwist rate at mitral valve opening of 57°/s, sensitivity 85% and specificity 48%; and degree of untwist at mitral valve opening of 51%, sensitivity 80%, specificity 98%. AUC ⫽ area under curve.
Figure 4. Averaged twist and untwist velocities throughout cardiac cycle related to controls and those with HC (time normalized by the length of systole and diastole, as appropriate). No statistically significant differences observed between both groups regarding values of peak untwist velocity (p ⫽ 0.3); however, timing of this event was significantly prolonged in patients with HC (p ⬍0.001).
more, in contrast to most 2D-STE studies,8 –10 our 3DSTE results (Figure 2) were notable for their consistency with the findings from magnetic resonance imaging studies, which have yielded data for PLVT of 16.3° to 19.9°.18,19 Additionally, we performed a comparison between 3D and 2D analysis of 20 cases from each group (Table 4). The main finding was the differential contribution of apical and basal rotations to LV wringing motion, a relevant aspect because of the crucial significance of the accurate assessment of regional mechanics. An explanation might be related to the elapsed time to rotational peak values; this issue is particularly relevant for 2D-STE, in which those values are calculated from 2 data sets, acquired at different times. In contrast, 3D-STE offers a different method, in which twist and its timing
are obtained over only 1 LV full volume. Other factors such as out-of-plane tissue movement12 and temporal resolution20 should also be taken into account. As with PLVT, enhanced contractility increases torsion recoil.17,21,22 However, HC seems not to follow that behavior, with a significantly low degree of untwist and untwist rate. The untwist at mitral valve opening was the parameter that best differentiated the patients with HC from the normal subjects (Figure 3). Similar conclusions were found in a study of patients with hypertension.23 Despite differences between HC and LV hypertrophy secondary to hypertension, these data might help to understand torsion recoil within the setting of an increased LV mass. The search for the main determinants of diastolic untwist has been cause for analysis,17,21,22 and it
1794
The American Journal of Cardiology (www.ajconline.org)
Figure 5. Scatter plots for PLVT and untwist rate at mitral valve opening in control and HC groups. Simple linear regression analysis showed statistically significant relation between untwist rate at mitral valve opening and PLVT values. With moderate degree of correlation in both controls (r ⫽ ⫺0.58) and patients with HC (r ⫽ ⫺0.52), greater PLVT associated with greater absolute untwist rate values. In both scatter plots, untwist rate at mitral valve opening refers to velocity of LV torsion recoil from its onset to mitral valve opening. Lines indicate regression line bounded by 95% CIs.
appears that torsion recoil relies primarily on parameters such as systolic twist.22 Although the present study lacked a multivariate analysis, our results favor those observations, showing that PLVT remained correlated with untwist rate in the HC group (Figure 5), despite the overall alteration of untwist. This finding supports the idea of twist as a phenomenon interrelating systolic contraction and diastolic relaxation.1– 4 Finally, PUV did not yield differences between the controls and the patients with HC, which might be interpreted, in the latter, as a compensatory mechanism aiding LV filling,22 occurring in the context of a low untwist rate and despite an increased PLVT. Another relevant aspect is the timing of PUV. Although a previous study reported no differences between patients with diastolic heart failure and healthy controls,22 we found this event delayed in the HC group (Figure 2). Our observation might have resulted from a more homogenous etiology of diastolic dysfunction and more marked hypertrophy.10,23 An analysis of HC according to outflow tract obstruction demonstrated increased basal rotation in those without obstruction, with significantly elevated twist compared with controls (Table 2). In contrast, patients with obstruction featured increased PLVT derived from both greater apical and basal rotation. These findings are consistent with a likely relation between a highly twisted left ventricle and outflow tract obstruction, with a lack of correlation between the pressure gradients and PLVT values, probably secondary to the influence of other components of myocardial mechanics.8,24 On the other hand, the untwist parameters did not differ according to the presence of obstruction (Table 3), suggesting the minor influence on torsion recoil that variations in LV afterload might have in these patients. Our study had several limitations. Eight patients were excluded because of poor image visualization. This issue highlights the dependence of this technique on the temporal– spatial resolution, which still precludes its widespread application. In addition, those patients were significantly older than the remaining subjects. This contributed to the young age of the HC group and limits the extrapolation of our values to an older population. Finally, we classified HC according to at rest or Valsalva-induced outflow tract obstruction, without considering the possible exercise-provocable gradient.
Table 4 Comparison between major parameters derived from 3-dimensional (3D) and 2-dimensional (2D) speckle tracking analysis (STE) Parameter
Peak apical rotation (°) Peak basal rotation (°) Peak left ventricular twist (°) Time to peak left ventricular twist* (%) Untwist rate at mitral valve opening (°/s) Untwist at mitral valve opening (%) * Time normalized by the length of systole.
Controls (n ⫽ 20)
HC (n ⫽ 20)
3D-STE
2D-STE
p Value
3D-STE
2D-STE
p Value
8.1 ⫾ 3.2 ⫺2.7 ⫾ 1.6 12.1 ⫾ 3.6 86 ⫾ 5 ⫺51 ⫾ 15 57 ⫾ 13
6.1 ⫾ 2.2 ⫺4.8 ⫾ 1.8 11.4 ⫾ 3.8 91 ⫾ 5 ⫺58 ⫾ 32 57 ⫾ 19
0.006 ⬍0.001 0.4 ⬍0.003 0.3 0.9
10.5 ⫾ 3.3 ⫺3.7 ⫾ 2.0 15.4 ⫾ 4.4 93 ⫾ 6 ⫺37 ⫾ 14 29 ⫾ 14
9.0 ⫾ 2.8 ⫺5.6 ⫾ 2.1 14.3 ⫾ 3.6 97 ⫾ 4 ⫺40 ⫾ 19 31 ⫾ 19
0.06 0.01 0.4 0.04 0.5 0.6
Cardiomyopathy/Twist Mechanics in Hypertrophic Cardiomyopathy 1. Notomi Y, Popovic ZB, Yamada H, Wallick DW, Martin MG, Oryszak SJ, Shiota T, Greenberg NL, Thomas JD. Ventricular untwisting: a temporal link between left ventricular relaxation and suction. Am J Physiol Heart Circ Physiol 2008;294:H505–H513. 2. 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. 3. Helmes M, Trombitás K, Granzier H. Titin develops restoring force in rat cardiac myocytes. Circ Res 1996;79:619 – 626. 4. Waldman LK, Nosan D, Villarreal F, Covell JW. Relation between transmural deformation and local myofiber direction in canine left ventricle. Circ Res 1988;63:550 –562. 5. Seo Y, Ishizu T, Enomoto Y, Sugimori H, Yamamoto M, Machino T, Kawamura R, Aonuma K. Validation of 3-dimensional speckle tracking imaging to quantify regional myocardial deformation. JACC Cardiovasc Imaging 2009;2:451– 459. 6. Sengupta PP, Tajik AJ, Chandrasekaran K, Khandheria BK. Twist mechanics of the left ventricle: principles and application. J Cardiovasc Coll Cardiol Imaging 2008;1:366 –376. 7. Geske JB, Sorajja P, Nishimura RA, Ommen SR. Evaluation of left ventricular filling pressures by Doppler echocardiography in patients with hypertrophic cardiomyopathy: correlation with direct left atrial pressure measurement at cardiac catheterization. Circulation 2007; 116:2702–2708. 8. Carasso S, Yang H, Woo A, Vannan MA, Jamorski M, Wigle ED, Rakowski H. Systolic myocardial mechanics in hypertrophic cardiomyopathy: novel concepts and implications for clinical status. J Am Soc Echocardiogr 2008;21:675– 683. 9. van Dalen BM, Kauer F, Soliman OI, Vletter WB, Michels M, ten Cate FJ, Geleijnse ML. Influence of the pattern of hypertrophy on left ventricular twist in hypertrophic cardiomyopathy. Heart 2009;95:657– 661. 10. Carasso S, Yang H, Woo A, Jamorski M, Wigle ED, Rakowski H. Diastolic myocardial mechanics in hypertrophic cardiomyopathy. J Am Soc Echocardiogr 2010;23:164 –171. 11. Chang SA, Kim HK, Kim DH, Kim JC, Kim YJ, Kim HC, Sohn DW, Oh BH, Park YB. Left ventricular twist mechanics in patients with apical hypertrophic cardiomyopathy: assessment with 2D speckle tracking echocardiography. Heart 2010;96:49 –55. 12. de Isla LP, Vivas D, Zamorano JL. Three-dimensional speckle tracking. Curr Cardiovasc Img Rep 2008;1:25–29. 13. Henry WL, Ware J, Gardin JM, Hepner SI, McKay J, Weiner M. Echocardiographic measurements in normal subjects: growth-related changes that occur between infancy and early adulthood. Circulation 1978:57:278 –285. 14. Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK, Pennell DJ, Rumberger JA, Ryan T, Verani MS; American Heart Association Writing Group on Myocardial Segmentation and Regis-
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
1795
tration for Cardiac Imaging. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: a statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation 2002;105:539 –542. Hansen DE, Daughters GT II, 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. 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. Dong SJ, Hees PS, Siu CO, Weiss JL, Shapiro EP. MRI assessment of LV relaxation by untwisting rate: a new isovolumic phase measure of tau. Am J Physiol Heart Circ Physiol 2001;281:H2002–H2009. 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. Young AA, Kramer CM, Ferrari VA, Axel L, Reichek N. Threedimensional left ventricular deformation in hypertrophic cardiomyopathy. Circulation 1994;90:854 – 867. Mor-Avi V, Lang RM, Badano LP, Belohlavek M, Cardim NM, Derumeaux G, Galderisi M, Marwick T, Nagueh SF, Sengupta PP, Sicari R, Smiseth OA, Smulevitz B, Takeuchi M, Thomas JD, Vannan M, Voigt JU, Zamorano JL. Current and evolving echocardiographic techniques for the quantitative evaluation of cardiac mechanics: ASE/ EAE consensus statement on methodology and indications endorsed by the Japanese Society of Echocardiography. J Am Soc Echocardiogr 2011;24:277–313. Moon MR, Ingels NB Jr, Daughters GT II, 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. Wang J, Khoury DS, Yue Y, Torre-Amione G, Nagueh SF. Left ventricular untwisting rate by speckle tracking echocardiography. Circulation 2007;116:2580 –2586. Takeuchi M, Borden WB, Nakai H, Nishikage T, Kokumai M, Nagakura T, Otani S, Lang RM. Reduced and delayed untwisting of the left ventricle in patients with hypertension and left ventricular hypertrophy: a study using two-dimensional speckle tracking imaging. Eur Heart J 2007;28:2756 –2762. van Ramshorst J, Mollema SA, Delgado V, van der Wall EE, Schalij MJ, Atsma DE, Bax JJ. Relation of immediate decrease in ventricular septal strain after alcohol septal ablation for obstructive hypertrophic cardiomyopathy to long-term reduction in left ventricular outflow tract pressure gradient. Am J Cardiol 2009;103:1592–1597.
From the Cardiovascular Imaging and Hemodynamic Laboratory, Tufts Medical Center, Boston Massachusetts. Manuscript received May 30, 2011; manuscript received and accepted July 20, 2011. The Cardiovascular Imaging and Hemodynamic Laboratory at Tufts Medical Center (Boston, Massachusetts) received an equipment grant from Toshiba Medical Systems (Tustin, California). *Corresponding author: Tel: 617-636-6151; fax: 617-636-8070. E-mail address: [email protected] (J. A. Urbano Moral). 0002-9149/11/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2011.07.047
interest in STE for the assessment of HC through rotational and twist parameters, the available data are quite scarce and limited to a few reports,8 –11 all of which used 2-dimensional technology, with the inherent limitation of tracking out-ofplane tissue motion.12 The aim of the present study was to describe LV twist mechanics in HC using 3-dimensional (3D)-STE. Methods We examined outpatients referred to our echocardiography laboratory from the Hypertrophic Cardiomyopathy Center (Tufts Medical Center, Boston, Massachusetts). Inclusion required a diagnosis of HC (the demonstration of a hypertrophied nondilated left ventricle [in adults, a wall thickness of ⱖ15 mm; in children, a wall thickness ⬎⫹2 SDs of normal values according to body surface area13]) and the absence of hypertension, diabetes mellitus, or any other cardiac or systemic disease that could produce the magnitude of hypertrophy evident. In addition, sinus rhythm had to be present at echocardiography. Left ventricular outflow tract obstruction was considered when an at rest or provocable (Valsalva maneuver) gradient of ⱖ30 mm Hg was detected. The patients were questioned in detail about their history and symptoms of coronary artery disease and were excluded if they provided any history suggesting this condition www.ajconline.org
Cardiomyopathy/Twist Mechanics in Hypertrophic Cardiomyopathy
or if evidence of coronary artery disease was identified by a review of the medical records. Patients demonstrating wall motion abnormalities on the standard echocardiogram were also excluded. In addition, as a part of the routine assessment of HC in our institution, 18 of 40 enrolled patients (those without at rest or Valsalva-induced obstruction) underwent a treadmill echocardiographic test to evaluate for exercise-provocable obstruction. All these patients reached the target heart rate, with neither ST-segment changes nor wall motion abnormalities on the peak exercise echocardiogram. Finally, patients with previous septal myectomy or alcohol septal ablation were excluded. Normal age- and gender-matched volunteers were examined as controls (⫾5 years matching for age and exact matching for gender). They were selected as normal subjects after a detailed history was obtained regarding possible health issues, with an absence of complaints compatible with cardiac disease and no history of hypertension, diabetes mellitus, or any other systemic disease. They were excluded as controls if any of the following was detected on the standard echocardiogram: significant valvulopathy; abnormal cardiac chamber size; or ejection fraction ⬍50%. The institutional review board approved the present study. Echocardiograms were performed with the Artida 4D System scanner (Toshiba Medical Systems, Tustin, California) in the tissue harmonic mode. The PST-30SBT transducer was used for standard 2D/Doppler measurements and the matrix array PST-25SX transducer for acquisition of 3D data sets. The latter consisted of an apical full volume created by the combination of 6 wedge-shaped subvolumes from 6 consecutive cardiac cycles during a single breathhold; 2 to 4 data sets with a temporal resolution of 20 to 25 volumes per second were obtained from each patient to subsequently select that of the best quality for off-line processing. The full-volume 3D data sets were analyzed using the 3D Wall Motion Tracking software (Toshiba Medical Systems) by an expert in the interpretation of echocardiographic images. First, the apical 4- and 2-chamber views and the 3 short axis views were automatically selected at end-diastole. Then, the foreshortened images were avoided by looking for the largest long axis dimensions. The transversal axes were modified as appropriate to optimize their orientation throughout the cardiac cycle. Next, the LV endocardial surface at end-diastole was automatically traced; only in the cases in which the software algorithm did not identify the surface correctly, manual tracing was used. Finally, myocardial tissue displacement was tracked throughout the cardiac cycle. The left ventricle was divided into 16 segments.14 The curves and values of LV rotation and twist and the timing of these parameters were provided by the software. Untwist and untwist rate were calculated by the investigators using the following formulas: Untwist ⫽ 关共 PLVT ⫺ Twist1兲 ⁄ PLVT兴 ⫻ 100 Untwist rate ⫽ 共Twist1 ⫺ PLVT兲 ⁄
关共t1 ⫺ tPLVT兲
⁄ 1000兴
where PLVT is the peak LV twist, Twist1 represents the degree of twist at a specific point along diastole, and t1 and tPLVT stand for the elapsed times at a specific point along
1789
Table 1 Baseline characteristics in patients with hypertrophic cardiomyopathy (HC) and controls Parameter Age (years) Women Heart rate (beats/min) Maximum left ventricular wall thickness (mm) Pattern of hypertrophy Septal asymmetric Symmetric (concentric) Predominantly midcavity New York Heart Association functional class Mitral regurgitation Number/trace Mild Moderate Severe Presence of left ventricular outflow tract obstruction Left ventricular outflow tract gradient at rest (mm Hg) History of paroxysmal atrial fibrillation Medications  Blockers Calcium channel blockers Disopyramide Diuretics
Controls (n ⫽ 40)
HC (n ⫽ 40)
p Value
35 ⫾ 10 17 (42.5%) 67 ⫾ 10 8⫾2
37 ⫾ 16 17 (42.5%) 66 ⫾ 12 21 ⫾ 4
0.5 1.0 0.7 ⬍0.001
35 (87.5%) 2 (5%) 3 (7.5%) 1.8 ⫾ 0.8
21 (52.5%) 12 (30%) 5 (12.5%) 2 (5%) 22 (55%) 42 ⫾ 38 7 (17.5%)
28 (70%) 27 (67.5%) 1 (2.5%) 1 (2.5%)
Continuous data are presented as mean ⫾ SD and categorical data are presented as n (%).
diastole and at PLVT, respectively. In our study, the untwist rate was calculated for mitral valve opening (difference in the degree of twist between mitral valve opening and peak LV twist divided into the corresponding interval). The denominator was divided by 1,000 to express the units in degrees/second. The instantaneous twist and untwist velocities were calculated through the difference in the degree of twist and time ratio between pairs of consecutive frames: Twist/untwist velocity ⫽ 共Twist1 ⫺ Twist0兲 ⁄
关共t1 ⫺ t0兲
⁄ 1000兴
The data are presented as the mean ⫾ SD or frequencies (percentages). An unpaired Student’s t test was used for comparisons between the normal subjects and those with HC, and between HC patients with and without outflow tract obstruction. The ability of the determined parameters to discriminate the patients with HC from the controls was evaluated by receiver operating characteristic curves. The optimal cutoff values were defined as those points providing the greatest sum of sensitivity and specificity. The role of the PLVT for predicting the untwist rate was assessed using simple linear regression analysis. The correlations were assessed using Pearson’s correlation coefficient, and the comparison between them by Fisher’s Z transformation. Statistical significance was defined by a 2-tailed p value of
1790
The American Journal of Cardiology (www.ajconline.org)
Table 2 Parameters derived from 3-dimensional speckle tracking analysis (3D-STE) Parameter End-diastolic volume (ml/m2) End-systolic volume (ml/m2) Ejection fraction (%) Left ventricular mass (g/m2) Peak apical rotation (°) Time to peak apical rotation§ (%) Peak basal rotation (°) Time to peak basal rotation§ (%) Peak left ventricular twist (°) Time to peak left ventricular twist§ (%) Peak twist velocity (°/s) Time to peak twist velocity§ (%)
Controls (n ⫽ 40)
All HC (n ⫽ 40)
p Value
Obstructive* HC (n ⫽ 22)
Nonobstructive* HC (n ⫽ 18)
p Value
62 ⫾ 9 25 ⫾ 9 59 ⫾ 3 73 ⫾ 11 8.2 ⫾ 3.6 83 ⫾ 8 ⫺2.6 ⫾ 1.6 87 ⫾ 7 12.0 ⫾ 3.9 85 ⫾ 6 62 ⫾ 19 57 ⫾ 13
62 ⫾ 13 23 ⫾ 5 63 ⫾ 4 104 ⫾ 26 11.3 ⫾ 4.0 91 ⫾ 8 ⫺3.7 ⫾ 1.9 92 ⫾ 10 16.5 ⫾ 4.7 94 ⫾ 6 77 ⫾ 19 57 ⫾ 19
0.9 0.03 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 0.005 0.007 ⬍0.001 ⬍0.001 0.001 0.9
63 ⫾ 13 23 ⫾ 6 63 ⫾ 4 109 ⫾ 28 12.7 ⫾ 4.4† 95 ⫾ 6 ⫺3.5 ⫾ 2.1¶ 90 ⫾ 10 18.4 ⫾ 4.6† 93 ⫾ 6 81 ⫾ 19 58 ⫾ 20
62 ⫾ 12 22 ⫾ 5 63 ⫾ 4 99 ⫾ 23 9.7 ⫾ 2.8‡ 88 ⫾ 10 ⫺3.9 ⫾ 1.5储 95 ⫾ 9 14.1 ⫾ 3.6# 94 ⫾ 7 72 ⫾ 18 56 ⫾ 17
0.8 0.7 0.9 0.3 0.02 0.01 0.6 0.08 0.003 0.9 0.2 0.8
Data are presented as mean ⫾ SD. * Left ventricular outflow tract obstruction and no left ventricular outflow tract obstruction. † p ⬍0.001 versus controls. ‡ p ⫽ 0.1 versus controls. ¶ p ⫽ 0.049 versus controls. 储 p ⫽ 0.005 versus controls. # p ⫽ 0.049 versus controls. § Time normalized by the length of systole.
⬍0.05. The inter- and intraobserver variability, tested by intraclass correlation coefficient, was as follows: PLVT 0.85 (95% confidence interval [CI] 0.69 to 0.94) and 0.89 (95% CI 0.71 to 0.97); untwist rate 0.88 (95% CI 0.72 to 0.96) and 0.90 (95% CI 0.72 to 0.97); and untwist at mitral valve opening 0.89 (95% CI 0.74 to 0.97) and 0.91 (95% CI 0.75 to 0.98). Results A total of 48 subjects with a diagnosis of HC were initially examined for eligibility for the present study. Inadequate visualization of the endocardial boundary precluded correct wall motion tracking in 8, who were eventually excluded from the analysis. The remaining 40 showed adequate tracking in all 16 LV segments and were consecutively included (Table 1). Up to 50 healthy volunteers (10 without appropriate wall motion tracking) were examined to form the control group. The 3D-STE parameters are summarized in Table 2. A systolic wringing motion with clockwise basal and counterclockwise apical rotation occurred in both groups. Peak LV twist showed significantly greater values in the HC group (16.5 ⫾ 4.7° vs 12.0 ⫾ 3.9°, p ⬍0.001). Figure 1 demonstrates examples of rotational and twist curves from the control and HC groups. The comparisons between HC patients with and without outflow tract obstruction showed differences in apical rotation and PLVT (Tables 2 and 3). In patients with obstruction, no significant correlations were found between those parameters and the at rest LV outflow tract gradient (apical rotation, r ⫽ 0.17, p ⫽ 0.4; PLVT, r ⫽ ⫺0.05, p ⫽ 0.8). Peak LV twist, and therefore, onset of LV untwist occurred later in those with HC (Table 2). From that point, untwist rate was slower in the HC group (Table 3), which was associated with only 31 ⫾ 12% of untwist at mitral
valve opening versus 62 ⫾ 15% in the control subjects (p ⬍0.001). Furthermore, the patients with HC expended 67 ⫾ 23% of diastole to reach 90% of untwist, an event that occurred at 45 ⫾ 21% of diastole in the controls (p ⬍0.001). Different parameter distributions and optimal cutoff values to discriminate those with HC from the controls are displayed in Figures 2 and 3. Although no significant difference was observed for peak untwist velocity (PUV), it occurred later during diastole in the HC group (Table 3 and Figure 4). Moreover, although the timing of PUV and mitral valve opening did not differ in the control group (13 ⫾ 9% vs 15 ⫾ 3%, p ⫽ 0.2; time normalized by the length of diastole), in the patients with HC, the former was significantly delayed with respect to isovolumic relaxation period (22 ⫾ 7% vs 17 ⫾ 4%, p ⬍0.001; time normalized by the length of diastole). The simple linear regression analysis found PLVT to be associated with the untwist rate at mitral valve opening in both controls and those with HC, with a moderate correlation between those parameters (Figure 5). No statistically significant difference was found between the respective correlation coefficients associated with the HC and control groups (p ⫽ 0.7). Discussion LV twist has a relevant role in maintaining efficient myocardial contraction and aids in generating suction power in early diastole.1– 4 The present study, using 3D-STE, allowed us to state the following observations: patients with HC showed increased PLVT that was mainly dependent on the apical rotation of those with LV outflow tract obstruction, onset of torsion recoil occurring close to the aortic valve closure, limited completion of untwist during early diastole, and delayed PUV. Initially, STE analysis of the patients with HC showed a mild rotation gradient between the LV apex and base.8
Cardiomyopathy/Twist Mechanics in Hypertrophic Cardiomyopathy
1791
Figure 1. Rotation (Top) and twist (Bottom) curves in control subject (A,C) and patient with HC (B,D). Apical and basal levels of patient with HC (B) showed greater peak rotation than those seen in control subject (A), evident in color-coded LV chamber schematics attached to each figure. In contrast, patient with HC featured greater PLVT (D), close to aortic valve closure. This was followed by lower degree of untwist at mitral valve opening compared to nearly complete untwist in control subject (C). AVC ⫽ aortic valve closure; MVO ⫽ mitral valve opening.
Subsequent reports suggested that this difference might be greater and especially influenced by the pattern of hypertrophy.9,11 In our HC group, most patients showed septal asymmetric hypertrophy involving primarily the mid and basal levels, points that should be considered when interpreting the prominent apical-basal rotation gradient found in these patients. In contrast, a greater apical and basal rotation in HC compared with healthy
controls has been reported by others.9 Our results, in keeping with those observations, suggest that although the basal ventricle is a major location for extreme hypertrophy, it maintains its contribution to the wringing motion during systole. In this regard, enhanced contractility has been demonstrated to markedly increase torsion15–17; therefore, the hypercontractile state in HC might explain the extended twist mechanics in these patients. Further-
1792
The American Journal of Cardiology (www.ajconline.org)
Table 3 Comparisons of untwist parameters in obstructive and nonobstructive hypertrophic cardiomyopathy (HC) and control groups Parameter
Controls (n ⫽ 40)
All HC (n ⫽ 40)
p Value
Obstructive* HC (n ⫽ 22)
Nonobstructive* HC (n ⫽ 18)
p Value
Isovolumic relaxation time (ms) Untwist at mitral valve opening (%) Untwist rate at mitral valve opening (°/s) Time to 90% of untwist† (%) Peak untwist velocity (°/s) Time to peak untwist velocity† (%) Time to mitral valve opening† (%)
79 ⫾ 8 62 ⫾ 15 ⫺56 ⫾ 23 45 ⫾ 21 ⫺78 ⫾ 27 13 ⫾ 9‡ 15 ⫾ 3
93 ⫾ 12 31 ⫾ 12 ⫺44 ⫾ 20 67 ⫾ 23 ⫺84 ⫾ 28 22 ⫾ 7§ 17 ⫾ 4
⬍0.001 ⬍0.001 0.01 ⬍0.001 0.3 ⬍0.001 0.008
95 ⫾ 12 31 ⫾ 12 ⫺48 ⫾ 21 72 ⫾ 21 ⫺90 ⫾ 31 22 ⫾ 7 17 ⫾ 3
91 ⫾ 13 31 ⫾ 12 ⫺39 ⫾ 17 60 ⫾ 25 ⫺77 ⫾ 24 23 ⫾ 7 17 ⫾ 5
0.4 0.9 0.2 0.1 0.2 0.7 0.9
Data are presented as mean ⫾ SD. * Left ventricular outflow tract obstruction and no left ventricular outflow tract obstruction. † Time normalized by the length of diastole. ‡ p ⫽ 0.2 versus time to mitral valve opening. § p ⬍0.001 versus time to mitral valve opening.
Figure 2. Distribution in peak LV twist, untwist rate at mitral valve opening, untwist at mitral valve opening, and peak untwist velocity related to controls and those with HC. Box plots defined by median (line in center of box), lower and upper quartiles (Q1 and Q3, holding 75% of central values in studied variable), minimum and maximum value within 1.5 interquartile range of lower and upper quartile, respectively (caps at end of each box), and values out of these limits (outliers, small circles).
Cardiomyopathy/Twist Mechanics in Hypertrophic Cardiomyopathy
1793
Figure 3. Receiver operating characteristic curves for those parameters with significant intergroup differences in Figure 2. Cutoff values that yielded greatest sensitivity and specificity for each parameter were as follows: PLVT of 13.3°, sensitivity 75% and specificity 73%; untwist rate at mitral valve opening of 57°/s, sensitivity 85% and specificity 48%; and degree of untwist at mitral valve opening of 51%, sensitivity 80%, specificity 98%. AUC ⫽ area under curve.
Figure 4. Averaged twist and untwist velocities throughout cardiac cycle related to controls and those with HC (time normalized by the length of systole and diastole, as appropriate). No statistically significant differences observed between both groups regarding values of peak untwist velocity (p ⫽ 0.3); however, timing of this event was significantly prolonged in patients with HC (p ⬍0.001).
more, in contrast to most 2D-STE studies,8 –10 our 3DSTE results (Figure 2) were notable for their consistency with the findings from magnetic resonance imaging studies, which have yielded data for PLVT of 16.3° to 19.9°.18,19 Additionally, we performed a comparison between 3D and 2D analysis of 20 cases from each group (Table 4). The main finding was the differential contribution of apical and basal rotations to LV wringing motion, a relevant aspect because of the crucial significance of the accurate assessment of regional mechanics. An explanation might be related to the elapsed time to rotational peak values; this issue is particularly relevant for 2D-STE, in which those values are calculated from 2 data sets, acquired at different times. In contrast, 3D-STE offers a different method, in which twist and its timing
are obtained over only 1 LV full volume. Other factors such as out-of-plane tissue movement12 and temporal resolution20 should also be taken into account. As with PLVT, enhanced contractility increases torsion recoil.17,21,22 However, HC seems not to follow that behavior, with a significantly low degree of untwist and untwist rate. The untwist at mitral valve opening was the parameter that best differentiated the patients with HC from the normal subjects (Figure 3). Similar conclusions were found in a study of patients with hypertension.23 Despite differences between HC and LV hypertrophy secondary to hypertension, these data might help to understand torsion recoil within the setting of an increased LV mass. The search for the main determinants of diastolic untwist has been cause for analysis,17,21,22 and it
1794
The American Journal of Cardiology (www.ajconline.org)
Figure 5. Scatter plots for PLVT and untwist rate at mitral valve opening in control and HC groups. Simple linear regression analysis showed statistically significant relation between untwist rate at mitral valve opening and PLVT values. With moderate degree of correlation in both controls (r ⫽ ⫺0.58) and patients with HC (r ⫽ ⫺0.52), greater PLVT associated with greater absolute untwist rate values. In both scatter plots, untwist rate at mitral valve opening refers to velocity of LV torsion recoil from its onset to mitral valve opening. Lines indicate regression line bounded by 95% CIs.
appears that torsion recoil relies primarily on parameters such as systolic twist.22 Although the present study lacked a multivariate analysis, our results favor those observations, showing that PLVT remained correlated with untwist rate in the HC group (Figure 5), despite the overall alteration of untwist. This finding supports the idea of twist as a phenomenon interrelating systolic contraction and diastolic relaxation.1– 4 Finally, PUV did not yield differences between the controls and the patients with HC, which might be interpreted, in the latter, as a compensatory mechanism aiding LV filling,22 occurring in the context of a low untwist rate and despite an increased PLVT. Another relevant aspect is the timing of PUV. Although a previous study reported no differences between patients with diastolic heart failure and healthy controls,22 we found this event delayed in the HC group (Figure 2). Our observation might have resulted from a more homogenous etiology of diastolic dysfunction and more marked hypertrophy.10,23 An analysis of HC according to outflow tract obstruction demonstrated increased basal rotation in those without obstruction, with significantly elevated twist compared with controls (Table 2). In contrast, patients with obstruction featured increased PLVT derived from both greater apical and basal rotation. These findings are consistent with a likely relation between a highly twisted left ventricle and outflow tract obstruction, with a lack of correlation between the pressure gradients and PLVT values, probably secondary to the influence of other components of myocardial mechanics.8,24 On the other hand, the untwist parameters did not differ according to the presence of obstruction (Table 3), suggesting the minor influence on torsion recoil that variations in LV afterload might have in these patients. Our study had several limitations. Eight patients were excluded because of poor image visualization. This issue highlights the dependence of this technique on the temporal– spatial resolution, which still precludes its widespread application. In addition, those patients were significantly older than the remaining subjects. This contributed to the young age of the HC group and limits the extrapolation of our values to an older population. Finally, we classified HC according to at rest or Valsalva-induced outflow tract obstruction, without considering the possible exercise-provocable gradient.
Table 4 Comparison between major parameters derived from 3-dimensional (3D) and 2-dimensional (2D) speckle tracking analysis (STE) Parameter
Peak apical rotation (°) Peak basal rotation (°) Peak left ventricular twist (°) Time to peak left ventricular twist* (%) Untwist rate at mitral valve opening (°/s) Untwist at mitral valve opening (%) * Time normalized by the length of systole.
Controls (n ⫽ 20)
HC (n ⫽ 20)
3D-STE
2D-STE
p Value
3D-STE
2D-STE
p Value
8.1 ⫾ 3.2 ⫺2.7 ⫾ 1.6 12.1 ⫾ 3.6 86 ⫾ 5 ⫺51 ⫾ 15 57 ⫾ 13
6.1 ⫾ 2.2 ⫺4.8 ⫾ 1.8 11.4 ⫾ 3.8 91 ⫾ 5 ⫺58 ⫾ 32 57 ⫾ 19
0.006 ⬍0.001 0.4 ⬍0.003 0.3 0.9
10.5 ⫾ 3.3 ⫺3.7 ⫾ 2.0 15.4 ⫾ 4.4 93 ⫾ 6 ⫺37 ⫾ 14 29 ⫾ 14
9.0 ⫾ 2.8 ⫺5.6 ⫾ 2.1 14.3 ⫾ 3.6 97 ⫾ 4 ⫺40 ⫾ 19 31 ⫾ 19
0.06 0.01 0.4 0.04 0.5 0.6
Cardiomyopathy/Twist Mechanics in Hypertrophic Cardiomyopathy 1. Notomi Y, Popovic ZB, Yamada H, Wallick DW, Martin MG, Oryszak SJ, Shiota T, Greenberg NL, Thomas JD. Ventricular untwisting: a temporal link between left ventricular relaxation and suction. Am J Physiol Heart Circ Physiol 2008;294:H505–H513. 2. 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. 3. Helmes M, Trombitás K, Granzier H. Titin develops restoring force in rat cardiac myocytes. Circ Res 1996;79:619 – 626. 4. Waldman LK, Nosan D, Villarreal F, Covell JW. Relation between transmural deformation and local myofiber direction in canine left ventricle. Circ Res 1988;63:550 –562. 5. Seo Y, Ishizu T, Enomoto Y, Sugimori H, Yamamoto M, Machino T, Kawamura R, Aonuma K. Validation of 3-dimensional speckle tracking imaging to quantify regional myocardial deformation. JACC Cardiovasc Imaging 2009;2:451– 459. 6. Sengupta PP, Tajik AJ, Chandrasekaran K, Khandheria BK. Twist mechanics of the left ventricle: principles and application. J Cardiovasc Coll Cardiol Imaging 2008;1:366 –376. 7. Geske JB, Sorajja P, Nishimura RA, Ommen SR. Evaluation of left ventricular filling pressures by Doppler echocardiography in patients with hypertrophic cardiomyopathy: correlation with direct left atrial pressure measurement at cardiac catheterization. Circulation 2007; 116:2702–2708. 8. Carasso S, Yang H, Woo A, Vannan MA, Jamorski M, Wigle ED, Rakowski H. Systolic myocardial mechanics in hypertrophic cardiomyopathy: novel concepts and implications for clinical status. J Am Soc Echocardiogr 2008;21:675– 683. 9. van Dalen BM, Kauer F, Soliman OI, Vletter WB, Michels M, ten Cate FJ, Geleijnse ML. Influence of the pattern of hypertrophy on left ventricular twist in hypertrophic cardiomyopathy. Heart 2009;95:657– 661. 10. Carasso S, Yang H, Woo A, Jamorski M, Wigle ED, Rakowski H. Diastolic myocardial mechanics in hypertrophic cardiomyopathy. J Am Soc Echocardiogr 2010;23:164 –171. 11. Chang SA, Kim HK, Kim DH, Kim JC, Kim YJ, Kim HC, Sohn DW, Oh BH, Park YB. Left ventricular twist mechanics in patients with apical hypertrophic cardiomyopathy: assessment with 2D speckle tracking echocardiography. Heart 2010;96:49 –55. 12. de Isla LP, Vivas D, Zamorano JL. Three-dimensional speckle tracking. Curr Cardiovasc Img Rep 2008;1:25–29. 13. Henry WL, Ware J, Gardin JM, Hepner SI, McKay J, Weiner M. Echocardiographic measurements in normal subjects: growth-related changes that occur between infancy and early adulthood. Circulation 1978:57:278 –285. 14. Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK, Pennell DJ, Rumberger JA, Ryan T, Verani MS; American Heart Association Writing Group on Myocardial Segmentation and Regis-
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
1795
tration for Cardiac Imaging. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: a statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation 2002;105:539 –542. Hansen DE, Daughters GT II, 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. 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. Dong SJ, Hees PS, Siu CO, Weiss JL, Shapiro EP. MRI assessment of LV relaxation by untwisting rate: a new isovolumic phase measure of tau. Am J Physiol Heart Circ Physiol 2001;281:H2002–H2009. 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. Young AA, Kramer CM, Ferrari VA, Axel L, Reichek N. Threedimensional left ventricular deformation in hypertrophic cardiomyopathy. Circulation 1994;90:854 – 867. Mor-Avi V, Lang RM, Badano LP, Belohlavek M, Cardim NM, Derumeaux G, Galderisi M, Marwick T, Nagueh SF, Sengupta PP, Sicari R, Smiseth OA, Smulevitz B, Takeuchi M, Thomas JD, Vannan M, Voigt JU, Zamorano JL. Current and evolving echocardiographic techniques for the quantitative evaluation of cardiac mechanics: ASE/ EAE consensus statement on methodology and indications endorsed by the Japanese Society of Echocardiography. J Am Soc Echocardiogr 2011;24:277–313. Moon MR, Ingels NB Jr, Daughters GT II, 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. Wang J, Khoury DS, Yue Y, Torre-Amione G, Nagueh SF. Left ventricular untwisting rate by speckle tracking echocardiography. Circulation 2007;116:2580 –2586. Takeuchi M, Borden WB, Nakai H, Nishikage T, Kokumai M, Nagakura T, Otani S, Lang RM. Reduced and delayed untwisting of the left ventricle in patients with hypertension and left ventricular hypertrophy: a study using two-dimensional speckle tracking imaging. Eur Heart J 2007;28:2756 –2762. van Ramshorst J, Mollema SA, Delgado V, van der Wall EE, Schalij MJ, Atsma DE, Bax JJ. Relation of immediate decrease in ventricular septal strain after alcohol septal ablation for obstructive hypertrophic cardiomyopathy to long-term reduction in left ventricular outflow tract pressure gradient. Am J Cardiol 2009;103:1592–1597.