Three-Dimensional Left Ventricular Torsion in Patients With Dilated Cardiomyopathy　― A Marker of Disease Severity ―

Abstract

Background: LV twist has a key role in maintaining left ventricular (LV) contractility during exercise. The purpose of this study was to investigate LV torsion instead of twist as a surrogate marker of peak oxygen uptake (peak V̇O₂) assessed by cardiopulmonary exercise testing (CPET) in patients with non-ischemic dilated cardiomyopathy (DCM).

Methods and Results: We evaluated 45 outpatients with DCM (50±12 years, 24% females) with 3D speckle-tracking electrocardiography prior to CPET. LV torsion, LV ejection fraction (EF), LV diastolic function, LV global longitudinal (GLS) and circumferential (GCS) strain were quantified. A reduced functional capacity (FC) was defined as a peak V̇O₂ <20 mL/kg/min. LV torsion correlated most strongly with peak V̇O₂ (r=0.76, P<0.001). LV torsion instead of twist was an independent predictor of peak V̇O₂ (B: 0.59 to 0.71, P<0.001) in multivariable analyses. Impaired LV torsion <0.61 degrees/cm was able to predict a reduced FC with higher sensitivity and specificity (0.91 and 0.81; area under the curve (AUC): 0.88, P<0.001) than LV EF, GLS or GCS (AUC 0.64, 0.63 and 0.66; P<0.05 for differences in AUC).

Conclusions: Peak V̇O₂ correlated more strongly with LV torsion than with LV diastolic function, LV EF, GLS or GCS. LV torsion had high accuracy in identifying patients with a reduced FC.

Idiopathic non-ischemic dilated cardiomyopathy (DCM) is characterized by progressive left ventricular (LV) dilatation that is accompanied by reduced myocardial contractility. Cardiopulmonary exercise testing (CPET), along with an abnormal peak oxygen uptake (peak V̇O₂ <20 mL/kg/min), is a strong predictor of death and disease progression in these patients.^1–3 In daily practice, therapy is most commonly assessed echocardiographically by changes in LV ejection fraction (EF). The validity of LV EF in predicting clinical outcome and peak V̇O₂ is questionable in chronic systolic heart failure.^4,5

Editorial p 440

LV twist, induced by systolic apical counterclockwise (positive angulation in echocardiography) and basal clockwise rotation, results in a LV wringing motion, which was first described in 1628 by Harvey.⁶ The counterdirectional helical layer architecture of the LV provides the anatomical basis of the twist motion.^7,8 A normalized twist, called torsion, makes measurements of the LV wringing motion more comparable among differently sized hearts. LV twist/torsion, together with longitudinal and circumferential deformation, indicates an effective LV performance, which is necessary for adequate stroke volume (SV) during incremental exercise.⁹ Twist motion and apical rotation are sensitive markers for changes in the state of cardiac function and contractility.^10,11

However, as with all 2D speckle-tracking techniques used for quantification of the LV, deformation and twist are affected by incorrect acquisition.¹² Cardiac motion is three-dimensional and myocardial speckles move in 3 dimensions. Hence, only a portion of the real myocardial motion can be detected with a 2D method. 3D speckle-tracking echocardiography (STE) is a novel tool to non-invasively quantify myocardial deformation and LV rotation. It has been validated against sonomicrometry, showing agreement with magnetic resonance imaging. 3D-STE allows efficient single-time acquisition of all LV segments in the same 3D data set.^13–16

Until now, no systematic studies have been able to elucidate the relationship between peak V̇O₂ and LV torsion in patients with DCM. The clinical relevance of LV torsion in treatment guidance for DCM is still unclear. Therefore, the aim of the present study was to analyse LV torsion as a potential marker for disease severity in DCM. 3D-STE may be helpful when CPET is contraindicated, physically impossible or inaccessible, especially in patients with advanced heart failure, and in a clinical or outpatient routine.

Methods

Study Population and Study Design

We prospectively screened consecutive ambulatory outpatients with recently diagnosed (>6 weeks but <3 months) DCM. The patients had a reduced LV EF <50% in the absence of coronary disease, and were receiving the best possible medical treatment according to current heart failure guidelines.¹⁷ All patients were in sinus rhythm, without ventricular pacing or cardiac resynchronization/modulation therapy. Exclusion criteria were as follows: (1) moderate or severe mitral regurgitation (n=7), (2) typical complete left bundle branch block (n=8), (3) inability to undergo CPET (n=3), and (4) poor acoustic window (n=6). All eligible patients underwent CPET with gas exchange analysis after echocardiography (within 8±5 days) without alteration of medical therapy. The study was approved by the local institutional review board and written informed consent was given by all participants.

3D and 2D Image Acquisition

Apical full 3D data sets of the LV were obtained using a PST-25SX (1–4 MHz) phased-array matrix transducer in the left lateral cubital position at rest (Toshiba Artida 4D System, Japan). Electrocardiogram-gated subvolumes of 4–6 heart cycles were acquired during a breathholding maneuvre. Elevation tilt, lateral contrast and volume rates (24±3 volumes per second) were adjusted for the best imaging quality. The 2D echocardiography was performed with the same ultrasound system immediately after 3D image acquisition. The pulsed-wave Doppler-derived early transmitral velocity (E) and tissue Doppler-derived lateral mitral annular velocity (E’) were obtained from the apical 4-chamber view. The E/E’ ratio was calculated to estimate LV filling pressure.¹⁸

3D Image Analysis

Offline analysis of the 3D data set was performed using commercial wall motion tracking software (WMT, Toshiba Medical System, Japan) as previously described¹⁹ (Figure 1). LV end-diastolic and end-systolic volumes as well as LV SV were indexed to body surface-area and used for the calculation of LV EF. The 3D LV sphericity index was calculated by dividing the measured LV end-diastolic volume by the volume of a sphere with a diameter derived from the end-diastolic LV axial length (a perfectly spherical LV would result in a value of 1.0). LV global longitudinal strain (GLS) and global circumferential strain (GCS) were calculated as the means of 16 regional peak systolic strain amplitudes.¹⁸ Rotation of the LV was performed in the same manner. Briefly, short-axis line markers were placed at the basal, mid-, and apical LV level, considering the papillary muscles as anatomic landmarks. The LV basal level was above the mitral annulus, the mid-level was across the papillary muscles, and the apical level was placed beyond the implantation of the papillary muscles. Rotation was computed by the software for each level (i.e., basal, mid- and apical levels) as the mean of the segments. LV twist was calculated as the systolic peak of the “instantaneous” difference between the basal and apical levels (Figure 1).²⁰ Rotation of the LV base and apex was expressed as peak systolic values (degrees). The software measured the distance between the basal and apical levels necessary to calculate LV torsion (twist per level distance=degrees/cm).

Figure 1.

Three-dimensional echocardiography of the left ventricular (LV) chamber (Upper). Apical 4- and 2-chamber views with corresponding basal, mid-level and apical cross-sections of the LV with results of rotation at each level. Exemplary comparison (Lower) of LV twist performance between 2 patients with reduced (Left) and preserved functional capacity (FC) (Right) with LV basal (dotted), apical (dashed) rotation and corresponding “instantaneous” twist curves (solid). Noteworthy between these patients are the accentuated differences in LV rotational mechanics, end-diastolic volume indices (EDV-I) and global circumferential strains (GCS) despite similar LV ejection fraction (EF) and global longitudinal strain (GLS).

Cardiopulmonary Exercise Testing

Patients underwent a symptom-limited bicycle CPET protocol with 3 distinct stages under medical supervision until exhaustion as previously described.²¹ Peak oxygen consumption (peak V̇O₂), minute ventilation/carbon dioxide production and the respiratory exchange ratio (RER) were measured or calculated respectively. Patients’ blood hemoglobin levels were measured before exercise. Peak RER was considered as a measure of exercise effort. Patients with peak RER ≤1.1 were excluded (n=3). Device calibration was performed before each test. Blood pressure values and 12-lead ECG were recorded continuously.

Statistical Analysis

Patients were divided into 2 groups according to the median value of the peak V̇O₂ distribution.

A reduced functional capacity (FC) was defined as peak V̇O₂ <20 mL/kg/min.¹ Results are expressed as mean±standard deviation or frequencies and percentages unless otherwise specified. The statistical differences between the groups were assessed using Student’s t-test, Wilcoxon’s test or Fisher’s exact test, as appropriate. The relationships among the echocardiographic and CPET data were evaluated using Pearson’s correlation coefficient r. To identify the independent predictors of peak V̇O₂, we used multiple linear regression analysis. We tested normality of distributions and heteroscedasticity as proposed.²² Only variables associated with peak V̇O₂ in the univariable analyses were included in the multilinear models. Effect sizes were expressed as standardized β estimates (B). Logistic regression analysis was applied for receiver-operating characteristic (ROC) with the area under the curve (AUC) and for calculation of the odds ratio (OR). ROC analysis was used to estimate the cut-off value of LV torsion, which was able to predict a reduced FC with corresponding sensitivity and specificity. Intra- and interobserver agreement was estimated from 21 randomly selected data sets within 1 week by 2 investigators blinded to previous results.²³ Statistical significance was defined as a two-tailed P value <0.05. All analyses were performed with open-source software R (version 3.0.2, 2013, The R Foundation for Statistical Computing, Vienna, Austria).

Results

Demographic and Clinical Characteristics of the Study Population

All 45 outpatients included in this study had clinically stable DCM without clinical signs of cardiac decompensation. Patients’ clinical and demographic characteristics and main CPET data are summarized in Table 1. None developed severe symptoms during CPET. The echocardiographic characteristics of the studied population are shown in Table 2.

Table 1. Demographic, Clinical and Cardiopulmonary Test Data for Study Patients With DCM

Variable	Overall (n=45)	Reduced FC (n=23)	Preserved FC (n=22)	P value
Demographic and clinical data
Age (years)	50±12	49±9	51±10	0.26
Female sex, n (%)	11 (24)	6 (26)	5 (23)	0.90
Body mass index (kg/m2)	25±4	25±4	25±4	0.58
Heart rate (beats/min)	73±8	73±6	73±9	0.91
SBP (mmHg)	127±19	126±18	128±19	0.82
DBP (mmHg)	74±15	73±17	75±13	0.91
NT-proBNP (pmol/L)	449 [151–1,471]	532 [155–1,234]	408 [141–944]	0.78
NYHA functional class	1.6±1.2	1.7±1.2	1.5±1.3	0.53
QRS duration (ms)	109±12	117±15	101±13	0.02
Medications
BB, n (%)	42 (93)	22 (96)	20 (91)	0.61
IAA, n (%)	43 (95)	22 (96)	21 (95)	0.99
Loop, n (%)	30 (66)	15 (65)	15 (68)	0.99
MRA, n (%)	32 (71)	17 (74)	15 (68)	0.74
Cardiopulmonary exercise test data
Peak workload (Watts)	142±44	123±32	155±45	0.02
Peak heart rate (beats/min)	135±28	134±29	138±26	0.59
Peak SBP (mmHg)	166±33	162±31	171±35	0.59
Peak DBP (mmHg)	86±16	89±18	83±14	0.36
V̇E/CO2 slope	34±6	36±6	31±6	0.04
Blood hemoglobin (mmol/L)	9.3±0.8	9.6±0.8	9.1±0.7	0.91

Data are presented as mean±standard deviation, absolute numbers n with relative frequencies (%) or as median with quantiles [25–75%] for NT-proBNP. P denotes the significance level of an unpaired t-test, χ²-test or Yuen’s modified t-test for NT-proBNP. BB, β-blockers; DCM, dilated cardiomyopathy; DBP, diastolic blood pressure; FC, functional capacity; IAA, inhibitors of angiotensin-converting enzyme and angiotensin receptor; Loop, loop diuretics; MRA, mineralocorticoid receptor antagonists; NT-proBNP, N-terminal pro B-type natriuretic peptide; NYHA, New York Heart Association; SBP, systolic blood pressure; V̇E/CO₂, minute ventilation/carbon dioxide production.

Table 2. Resting Echocardiographic Data for Study Patients With DCM

Variable	Overall (n=45)	Reduced FC (n=23)	Preserved FC (n=22)	P value
LV geometry
EDV-index (mL/m2)	101±27	109±31	92±19	0.03
Sphericity	0.67±0.07	0.70±0.08	0.65±0.07	0.05
ESV-index (mL/m2)	69±24	78±24	61±17	0.03
LV function
LV EF (%)	30±8	29±8	31±8	0.34
SV-index (mL/m2)	32±9	31±10	33±8	0.46
CI (mL/m2/min)	2.5±0.8	2.5±0.8	2.6±0.5	0.46
E/E’ ratio	15±3	16±2	13±2	<0.001

Data are presented as mean±standard deviation. P denotes the significance level of an unpaired t-test. CI, cardiac index; EDV, end-diastolic volume; E/E’, ratio of pulsed-wave Doppler-derived early diastolic transmitral flow and tissue Doppler-derived early diastolic velocity from the lateral mitral annulus; EF, ejection fraction; ESV, end-systolic volume; LV, left ventricular; SV, stroke volume. Other abbreviations as in Table 1.

Functional Capacity and DCM

Overall peak V̇O₂ had a mean of 20.2±5.7 mL/kg/min (median 19.9 mL/kg/min; range 11.9–33.4 mL/kg/min); 23 patients (51 %) had a reduced FC (peak V̇O₂ <20 mL/kg/min) with longer QRS duration, lower peak workload (Table 1), more dilated LV and a higher degree of diastolic dysfunction (Table 2). Patients with a reduced FC had significantly less LV twist motion, often with more reversed apical rotation (<0 degrees) than patients with a preserved FC (Table 3). GLS was statistically similar in the 2 groups, although there was a trend of lower GCS in patients with a reduced FC (Table 3). GCS and LV spherical dilation were significantly correlated (r=0.60, P<0.001). Age, sex, LV EF, cardiac index and medical therapy regimens were similar in the 2 groups (Table 1). There was a significant correlation between absolute peak V̇O₂ and the 2 following clinical demographic variables: N-terminal pro-B-type peptide and QRS duration (r=−0.33 and −0.36, P<0.05 for both) (Table 4).

Table 3. LV Deformation Parameters in Study Patients With DCM

Variable	Overall (n=45)	Reduced FC (n=23)	Preserved FC (n=22)	P value
LV strain
GLS (%)	−8.9±3.1	−8.2±3.5	−9.3±2.6	0.18
GCS (%)	−12.8±4.6	−11.6±4.6	−14.1±4.3	0.07
LV twist motion and rotation
Twist (degrees)	4.3±2.0	3.1±1.4	5.5±1.7	<0.001
Torsion (degrees/cm)	0.61±0.18	0.42±0.17	0.81±0.19	<0.001
Apical rotation (degrees)	1.4±2.1	0.7±2.1	2.1±2.1	0.02
Reverse apical rotation, n (%)	10 (23)	9 (39)	1 (4)	0.001
Basal rotation (degrees)	−3.2±2.0	−2.6±1.6	−3.7±2.2	0.08

Data are presented as mean±standard deviation, absolute numbers n with relative frequencies (%). P denotes the significance level of an unpaired t-test. GCS, global circumferential strain; GLS, global longitudinal strain; reverse apical rotation, <0 degrees. Other abbreviations as in Tables 1,2.

Table 4. Relationship Among Peak V̇O₂ and the Demographic, Clinical, and Echocardiographic Parameters in the Study Patients With DCM

Variables	r	P value
Demographic and clinical data
Age	0.07	0.63
Body mass index	−0.02	0.91
NT-proBNP (log)	−0.33	0.04
QRS duration	−0.36	0.01
LV geometry and function
EDV-index	−0.43	0.003
Sphericity	−0.31	0.05
ESV-index	−0.38	0.009
EF	0.36	0.02
SV-index	0.02	0.87
E/E’ ratio	−0.56	<0.001
LV deformation parameters
GLS	−0.36	0.03
GCS	−0.39	0.01
Twist	0.68	<0.001
Torsion	0.76	<0.001
Apical rotation	0.49	<0.001
Basal rotation	−0.26	0.09

Data are presented as correlation coefficient r with corresponding significance level P value. Peak V̇O₂, peak oxygen uptake; NT-proBNP, N-terminal pro brain natriuretic peptide - log transformed. Other abbreviations as in Tables 1–3.

Relationship Between LV Function Parameters and Functional Capacity

Among the LV function parameters, torsion, twist, diastolic function and apical rotation were the strongest univariate determinants of peak V̇O₂ (r=0.76, 0.68, 0.56 and 0.49, P<0.001 for all) (Figure 2, Table 4). However, in the multivariable analyses with adjustments for patients’ demographic and clinical parameters and LV function, peak LV torsion was an independent predictor of peak V̇O₂ (B=0.59–0.71, P<0.001 for all) (Table 5). Moreover, LV apical rotation was more strongly associated with LV torsion (r=0.61) than with basal rotation (r=−0.42) (P<0.01 for both). LV torsion significantly correlated with the E/E’ ratio, peak systolic blood pressure, peak workload and QRS duration (r=−0.57, 0.45, 0.46 and −0.40, respectively; P<0.05 for all). Furthermore, LV torsion was significantly associated with LV dilation as with LV sphericity and LV end-systolic volume (r=0.35 and −0.42, respectively; P<0.05 for both). As anticipated, LV twist (i.e., “unadjusted raw torsion”) was the main determinant of peak V̇O₂ (B=0.50–0.67, P<0.01) with QRS duration, LV end-diastolic volume and LV E/E’ ratio as significant covariates (B=−0.24, −0.33 and −0.29, respectively; P<0.03 for all) (Table S1). In addition, LV twist and torsion correlated significantly with GCS (r=−0.45 and −0.41, respectively; P<0.005 for both), but no associations were found with GLS (r=−0.27 and −0.29, respectively).

Figure 2.

Strong relationships of left ventricular (LV) torsion with peak oxygen uptake (peak V̇O₂) and diastolic dysfunction (E/E’ratio) in the total study population (n=45) with regression line (solid), corresponding 95% confidence interval (grey area), correlation coefficient r and corresponding significance level P; E/E’=ratio of pulsed-wave Doppler-derived early diastolic transmitral flow and tissue Doppler-derived early diastolic velocity from the lateral mitral annulus.

Table 5. Multilinear Regression of Torsion for Prediction of Peak V̇O₂ in Study Patients With DCM

	B	SE	P value
Model 1: demographic and clinical factors (r2=0.56, P<0.001)
Torsion	0.69	0.10	<0.001
NT-proBNP (log)	−0.16	0.11	0.13
QRS duration	−0.16	0.10	0.12
Model 2: LV geometry (r2=0.58, P<0.001)
Torsion	0.71	0.10	<0.001
EDV-index	−0.18	0.2	0.13
Sphericity	0.11	0.11	0.92
Model 3: LV function and deformation (r2=0.58, P<0.001)
Torsion	0.59	0.13	<0.001
E/E’ ratio	−0.22	0.12	0.07
EF	−0.40	0.39	0.31
GCS	−0.42	0.32	0.20
GLS	−0.13	0.20	0.53

B, standardised β-estimate; P, corresponding significance level; r², adjusted regression coefficient of multilinear regression model; SE, standard error of estimate. Other abbreviations as in Tables 1–3.

Echocardiographic Predictors of Reduced Functional Capacity

Using ROC analysis, LV torsion had a sensitivity of 0.91 and a specificity of 0.81 (AUC=0.88, P<0.001) to detect a reduced FC with a best cut-off value of 0.60 degrees/cm (OR=0.05, 95% confidence interval 0.01–0.19, P<0.001). Further results of the ROC analysis are depicted in Figure 3.

Figure 3.

Receiver-operating characteristic curve analyses showing that left ventricular twist is more accurate than other myocardial strain and functional parameters in identifying patients with a peak V̇O₂ of 20 mL/kg/min. *DeLong’s test (Left): comparison of differences between curves with P<0.05. AUC, area under the curve; ESV, end-systolic volume. Other abbreviations as in Figures 1,2.

Intra- and Interobserver Agreement of 3D-STE Parameters

Intra- and interobserver agreements were good for all LV deformation parameters: (a) intra-observer: LV apical rotation 94%; LV basal rotation 94%, LV GLS 96% and LV GCS 93%; (b) interobserver: LV apical rotation 92%; LV basal rotation 93%, LV GLS 94% and LV GCS 93%.

Discussion

The present study evaluated the relationship between LV twist and FC in patients with DCM and tested LV torsion as a surrogate parameter for FC. The major findings of this study are as follows. (1) LV twist, torsion and the E/E’ ratio were more closely correlated to peak V̇O₂ than LV EF or other strain parameters (Figure 2, Table 4). (2) Patients with a reduced FC (peak V̇O₂ <20 mL/kg/min) often had more severely reduced LV twisting motion with reversed apical rotation (Table 3). (3) LV torsion was an important determinant of a reduced FC (Figure 3); a cut-off value of 0.61 degrees/cm could predict a reduced FC with high sensitivity and specificity in patients with DCM.

LV Torsion and Peak V̇O₂

Peak V̇O₂ represents peak exercise cardiac output, which in turn relies on an adaptive LV SV. This adaptive mechanism is primarily dependant on the performance of systolic and diastolic coupling of a twist motion during incremental exercise.^9,24 In the present study, we demonstrated for the first time a strong positive correlation of LV twist and torsion with peak V̇O₂ (Table 4, Figure 2). The LVs, particularly in DCM, exhibit a progressive spherical dilatation with stretching of the myofibers in a more transverse orientation.²⁵ After an initial phase of reduced LV longitudinal function, LV circumferential strain and the twist motion further deteriorate during progression of heart failure.^20,26,27 Indeed, we showed a relationship between LV dilation and twist (r=−0.41) negatively affecting peak V̇O₂ (Table S1) in patients with a moderately to severely reduced LV EF. In other words, twist is dependent on the magnitude of LV dilation, which influences global systolic function.²⁸ At first glance, it may appear that LV torsion is independent of LV size for prediction of FC (Table 5), but torsion represents the net twist angle normalized by the LV long-axis length. It anticipates, therefore, the amount of LV dilation in progressive heart failure.

At this stage of the disease, torsional dysfunction is associated with impairment of circumferential strain (r=−0.41), but not with longitudinal deformation. Interestingly, Hasselberg and colleagues were able to establish an association between longitudinal strain and peak V̇O₂ in heart failure with preserved LV EF, but not in patients with a reduced LV EF.²¹ Recent studies have reported that LV twist in patients with systolic heart failure (overall mean LV EF 31%) is reduced within a range of 4.1–6.1 degrees compared with healthy persons.^29–33 The present result, 4.3±2.0 degrees of LV twist, is consistent with those reports, irrespective of the imaging method.

LV Torsion and Disease Severity

FC with peak V̇O₂ is a strong clinical indicator of disease progression and of survival in heart failure.³ We showed that patients with inappropriate twisting motion, dilated LVs, increased diastolic dysfunction and prolonged QRS duration adapt inadequately during exercise (Tables 1,3). Impairment of GLS and LV EF were not related to a reduced FC (peak V̇O₂ <20 mL/kg/min) (Table 3). In the progression of DCM, mid-myocardial replacement fibrosis disturbs circumferential deformation and twist motion, which results in a “stiffer” LV and leads to deleterious systolic and diastolic ventricular function.³³ This corroborates our finding that LV torsion is more appropriate than longitudinal strain for determining a reduced FC (Figure 3). Furthermore, we demonstrated that LV torsion was strongly associated with the amount of diastolic dysfunction (r=−0.67) in DCM patients.

LV diastolic function plays an important role in LV contractility through effective coupling of ventricular blood suction and ejection.³⁴ In contrast, we demonstrated clearly that LV torsion is an independent predictor of peak V̇O₂ in DCM patients with moderately to severely reduced LV EF (Table 5). Our results suggested that LV torsion not only epitomizes solely systolic LV function, but also represents a link to diastolic function in DCM. However, the eventual mechanism linking fibrosis and disease progression needs further exploration. Popescu and colleagues demonstrated that DCM patients with a reversed apical rotation and blunted twist motion exhibit severe LV remodelling compared with those with normally directed apical rotation.³⁵ In line with these observations, we found that patients with a reduced FC often had a more reversed LV apical rotation (i.e., negative angulation) (Figure 2, Table 2) with prolonged QRS duration (Table 3). A prolonged QRS duration is indicative of a higher degree of electrical dyssynchrony. Indeed, left intraventricular dyssynchrony triggers unfavourable twist mechanics and negatively affects functional capacity.³⁶ The amplitude of LV twist motion and peak V̇O₂ are inversely related to QRS duration even in the absence of a left bundle branch block as demonstrated in our study. Recent studies underpinned the use of cardiac resynchronization therapy on LV twist and on LV reverse remodelling in systolic heart failure patients.^30,31

Clinical Implications

The results of the present study demonstrated a clear relationship between LV torsion and functional capacity, suggesting that LV torsion may have strong prognostic relevance in DCM patients. Echocardiography represents the basic examination for initial diagnosis and follow-up of patients with heart failure. The results of this study support the idea that twist assessment should be more prominent in first-line and follow-up examinations of heart failure patients, avoiding repeated CPET in unstable or physically disabled persons. STE might be a valuable, easily accessible and non-invasive tool in the diagnostic work-up of DCM patients to monitor subtle changes in disease progression and therapy. Nevertheless, more studies are needed to elucidate the clinical effect of LV twist in DCM.

Study Limitations

Our study was designed to test whether LV torsion can function as a surrogate parameter of peak V̇O₂ in DCM patients. The limited number of patients did not allow us to provide a definitive statement about the clinical outcome. The degree of torsion is age-dependent, which might limit the generalisability of our study results to a broader population of heart failure patients. The prognostic value of peak V̇O₂ between 15 and 20 mL/kg/min is debatable. However, we used an absolute peak V̇O₂ value of 20 mL/kg/min in our study as recommended to assess heart failure severity and prognosis. 3D echocardiography is limited by its relatively low volume rate, which decreases its ability to capture events occurring in the fast phases of the cardiac cycle, such as isovolumetric relaxation. Hence, we have not assessed LV untwist, which measures the diastolic LV reverse rotation during isovolumetric relaxation time. However, we demonstrated good agreement of the speckle-tracking system in measuring systolic rotation. The fact that we demonstrated a relationship between functional capacity and resting myocardial function strengthens the clinical importance of the twist motion. Nevertheless, we had no echocardiographic information during exercise. Our findings should be confirmed in a larger population of patients to provide a direct comparison between 3D echocardiographic measures, exercise capacity and clinical outcome.

Conclusions

Our study has provided novel data showing that LV torsion was highly correlated to functional capacity and LV function. Furthermore, LV torsion was superior to LV EF in identifying patients with reduced FC. A threshold of 0.61 degrees/cm in LV torsion was highly sensitive and specific in predicting impaired cardiopulmonary fitness. Considering the strong relationship between FC and cardiac prognosis, LV torsion derived from 3D echocardiography may have the potential to identify and follow-up patients with a poorer prognosis and may open the potential for a new clinical routine in heart failure patients.

Grants

This research received no grants from any funding agency in the public, commercial or not-for-profit sectors.

Disclosures

The authors declare no conflicts of interest.

Supplementary Files

Supplementary File 1

Table S1. Multilinear regression analysis of twist for prediction of peak V̇O₂ in study patients with DCM

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-16-0965

References

• 1.   Guazzi M, Adams V, Conraads V, Halle M, Mezzani A, Vanhees L, et al. Clinical recommendations for cardiopulmonary exercise testing data assessment in specific patient populations. Eur Heart J 2012; 33: 2917–2927.
• 2.   Weber KT, Kinasewitz GT, Janicki JS, Fishman AP. Oxygen utilization and ventilation during exercise in patients with chronic cardiac failure. Circulation 1982; 65: 1213–1223.
• 3.   Francis DP, Shamim W, Davies LC, Piepoli MF, Ponikowski P, Anker SD, et al. Cardiopulmonary exercise testing for prognosis in chronic heart failure: Continuous and independent prognostic value from VE/VCO2 slope and peak VO2. Eur Heart J 2000; 21: 154–161.
• 4.   Sengelov M, Jorgensen PG, Jensen JS, Bruun NE, Olsen FJ, Fritz-hansen T, et al. Global longitudinal strain is a superior predictor of all-cause mortality in heart failure with reduced ejection fraction. JACC Cardiovasc Imaging 2015; 8: 1351–1359.
• 5.   Zhang KW, French B, May Khan A, Plappert T, Fang JC, Sweitzer NK, et al. Strain improves risk prediction beyond ejection fraction in chronic systolic heart failure. J Am Heart Assoc 2014; 3: e000550–e000550.
• 6.   Weil E. William Fitzer, the publisher of Harvey’s De motu cordis, 1628. Library (Lond), 1944; s4-XXIV(3–4): 142–164.
• 7.   Buckberg GD, Hoffman JI, Coghlan HC, Nanda NC. Ventricular structure-function relations in health and disease. Part I: The normal heart. Eur J Cardiothorac Surg 2014; 47: 1–15.
• 8.   Torrent-Guasp F, Buckberg GD, Clemente C, Cox JL, Coghlan HC, Gharib M. The structure and function of the helical heart and its buttress wrapping. I: The normal macroscopic structure of the heart. Semin Thorac Cardiovasc Surg 2001; 13: 301–319.
• 9.   Doucende G, Schuster I, Rupp T, Startun A, Dauzat M, Obert P, et al. Kinetics of left ventricular strains and torsion during incremental exercise in healthy subjects: The key role of torsional mechanics for systolic-diastolic coupling. Circ Cardiovasc Imaging 2010; 3: 586–594.
• 10.   Kanzaki H, Nakatani S, Yamada N, Urayama SI, Miyatake K, Kitakaze M. Impaired Systolic torsion in dilated cardiomyopathy: Reversal of apical rotation at mid-systole characterized with magnetic resonance tagging method. Basic Res Cardiol 2006; 101: 465–470.
• 11.   Kim WJ, Lee BH, Kim YJ, Kang JH, Jung YJ, Song JM, et al. Apical rotation assessed by speckle-tracking echocardiography as an index of global left ventricular contractility. Circ Cardiovasc Imaging 2009; 2: 123–131.
• 12.   Van Dalen BM, Vletter WB, Soliman OII, ten Cate FJ, Geleijnse ML. Importance of transducer position in the assessment of apical rotation by speckle tracking echocardiography. J Am Soc Echocardiogr 2008; 21: 895–898.
• 13.   Zhou Z, Ashraf M, Hu D, Dai X, Xu Y, Kenny B, et al. Three-dimensional speckle-tracking imaging for left ventricular rotation measurement: An in vitro validation study. J Ultrasound Med 2010; 29: 903–909.
• 14.   Maffessanti F, Nesser HJ, Weinert L, Steringer-Mascherbauer R, Niel J, Gorissen W, et al. Quantitative evaluation of regional left ventricular function using three-dimensional speckle tracking echocardiography in patients with and without heart disease. Am J Cardiol 2009; 104: 1755–1762.
• 15.   Wu VCC, Takeuchi M, Otani K, Haruki N, Yoshitani H, Tamura M, et al. Effect of through-plane and twisting motion on left ventricular strain calculation: Direct comparison between two-dimensional and three-dimensional spe ckle-tracking echocardiography. J Am Soc Echocardiogr 2013; 26: 12–14.
• 16.   Ashraf M, Zhou Z, Nguyen T, Ashraf S, Sahn DJ. Apex to base left ventricular twist mechanics computed from high frame rate two-dimensional and three-dimensional echocardiography: A comparison study. J Am Soc Echocardiogr 2012; 25: 121–128.
• 17.   McMurray JJ V, Adamopoulos S, Anker SD, Auricchio A, Böhm M, Dickstein K, et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur J Heart Fail 2012; 14: 803–869.
• 18.   Evangelista A, Flachskampf F, Lancellotti P, Badano L, Aguilar R, Monaghan M, et al. European Association of Echocardiography recommendations for standardization of performance, digital storage and reporting of echocardiographic studies. Eur J Echocardiogr 2008; 9: 438–448.
• 19.   Urbano-Moral JA, Arias-Godinez JA, Ahmad R, Malik R, Kiernan MS, DeNofrio D, et al. Evaluation of myocardial mechanics with three-dimensional speckle tracking echocardiography in heart transplant recipients: Comparison with two-dimensional speckle tracking and relationship with clinical variables. Eur Heart J Cardiovasc Imaging 2013; 14: 1167–1173.
• 20.   Sengupta PP, Tajik J, Chandrasekaran K, Khandheria BK. Twist mechanics of the left ventricle: Principles and application. JACC Cardiovasc Imaging 2008; 1: 366–376.
• 21.   Hasselberg NE, Hauga KH, Sarvari SI, Gullestad L, Andreassen K, Smiseth O, et al. Left ventricular global longitudinal strain is associated with exercise capacity in failing hearts with preserved and reduced ejection fraction. Eur Heart J Cardiovasc Imaging 2015; 16: 217–224.
• 22.   Crawford SL. Correlation and regression. Circulation 2006 114: 2083–2088.
• 23.   Kottner J, Audigé L, Brorson S, Donner A, Gajewski BJ, Hróbjartsson A, et al. Guidelines for Reporting Reliability and Agreement Studies (GRRAS) were proposed. J Clin Epidemiol 2011 64: 96–106.
• 24.   Stöhr EJ, González-Alonso J, Shave R. Left ventricular mechanical limitations to stroke volume in healthy humans during incremental exercise. Am J Physiol Heart Circ Physiol 2011; 301: H478–H487.
• 25.   Buckberg GD, Hoffman JIE, Coghlan HC, Nanda NC. Ventricular structure-function relations in health and disease. Part II: Clinical considerations. Eur J Cardio-Thoracic Surg 2014; 47: 778–787.
• 26.   Taber LA, Yang M, Podszus WW. Mechanics of ventricular torsion. J Biomech 1996; 29: 745–752.
• 27.   Zhang X, Haynes P, Campbell KS, Wenk JF. Numerical evaluation of myofiber orientation and transmural contractile strength on left ventricular function. J Biomech Eng 2015; 137: 044502.
• 28.   Van Dalen BM, Kauer F, Vletter WB, Soliman OII, van der Zwaan HB, Ten Cate FJ, et al. Influence of cardiac shape on left ventricular twist. J Appl Physiol 2010; 108: 146–151.
• 29.   Chang SN, Lai YH, Yen CH, Tsai CT, Lin JW, Bulwer BE, et al. Cardiac mechanics and ventricular twist by three-dimensional strain analysis in relation to B-type natriuretic peptide as a clinical prognosticator for heart failure patients. PLoS One 2014; 9: e115260.
• 30.   Bertini M, Marsan NA, Delgado V, van Bommel RJ, Nucifora G, Borleffs CJW, et al. Effects of cardiac resynchronization therapy on left ventricular twist. J Am Coll Cardiol 2009; 54: 1317–1325.
• 31.   Matsumoto K, Tanaka H, Tatsumi K, Miyoshi T, Hiraishi M, Kaneko A, et al. Left ventricular dyssynchrony using three-dimensional speckle-tracking imaging as a determinant of torsional mechanics in patients with idiopathic dilated cardiomyopathy. Am J Cardiol 2012; 109: 1197–1205.
• 32.   Saito M, Okayama H, Nishimura K, Ogimoto A, Ohtsuka T, Inoue K, et al. Determinants of left ventricular untwisting behaviour in patients with dilated cardiomyopathy: Analysis by two-dimensional speckle tracking. Heart 2009; 95: 290–296.
• 33.   Taylor RJ, Umar F, Lin ELS, Ahmed A, Moody WE, Mazur W, et al. Mechanical effects of left ventricular midwall fibrosis in non-ischemic cardiomyopathy. J Cardiovasc Magn Reson 2016; 18: 1–8.
• 34.   Notomi Y, Popovic ZB, Yamada H, Wallick DW, Martin MG, Oryszak SJ, et al. Ventricular untwisting: A temporal link between left ventricular relaxation and suction. Am J Physiol Heart Circ Physiol 2008; 294: H505–H513.
• 35.   Popescu BA, Beladan CC, Calin A, Muraru D, Deleanu D, Rosca M, et al. Left ventricular remodelling and torsional dynamics in dilated cardiomyopathy: Reversed apical rotation as a marker of disease severity. Eur J Heart Fail 2009; 11: 945–951.
• 36.   Tan YT, Wenzelburger FW, Sanderson JE, Leyva F. Exercise-induced torsional dyssynchrony relates to impaired functional capacity in patients with heart failure and normal ejection fraction. Heart 2013; 99: 259–266.