- Email: [email protected]
Utility of Comprehensive Assessment of Strain Dyssynchrony Index by Speckle Tracking Imaging for Predicting Response to Cardiac Resynchronization Therapy
Utility of Comprehensive Assessment of Strain Dyssynchrony Index by Speckle Tracking Imaging for Predicting Response to Cardiac Resynchronization Therapy Kazuhiro Tatsumi, MDa, Hidekazu Tanaka, MD, PhDa,*, Kouhei Yamawaki, MDa, Keiko Ryo, MDa, Alaa Mabrouk Salem Omar, MDa, Yuko Fukuda, MDa, Kazuko Norisada, MD, PhDa, Kensuke Matsumoto, MDa, Tetsuari Onishi, MD, PhDa, John Gorcsan III, MDb, Akihiro Yoshida, MD, PhDa, Hiroya Kawai, MD, PhDa, and Ken-ichi Hirata, MD, PhDa The strain delay index is reportedly a marker of dyssynchrony and residual myocardial contractility. The aim of this study was to test the hypothesis that a relatively simple version of the strain dyssynchrony index (SDI) can predict response to cardiac resynchronization therapy (CRT) and that combining assessment of radial, circumferential, and longitudinal SDI can further improve the prediction of responders. A total of 52 patients who underwent CRT were studied. The SDI was calculated as the average difference between peak and end-systolic strain from 6 segments for radial and circumferential SDI and 18 segments for longitudinal SDI. Conventional dyssynchrony measures were assessed by interventricular mechanical delay, the Yu index, and radial dyssynchrony by speckle tracking strain. Response was defined as a >15% decrease in end-systolic volume after 3 months. Of the individual parameters, radial SDI >6.5% was the best predictor of response to CRT, with sensitivity of 81%, specificity of 81%, and an area under the curve of 0.87 (p <0.001). Circumferential SDI >3.2% and longitudinal SDI >3.6% were also found to be predictive of response to CRT, with areas under the curve of 0.81 and 0.80, respectively (p <0.001). Moreover, radial, circumferential, and longitudinal SDI at baseline were correlated with reduction of end-systolic volume with CRT. In addition, the response rate in patients with 3 positive SDIs was 100%. In contrast, rates in patients with either 1 or no positive SDIs were 42% and 22%, respectively (p <0.005 and p <0.001 vs 3 positive SDIs). In conclusion, the SDI can successfully predict response to CRT, and the combined approach leads to more accurate prediction than using individual parameters. © 2011 Elsevier Inc. All rights reserved. (Am J Cardiol 2011;107:439 – 446) Because roughly 1/3 of patients do not respond to cardiac resynchronization therapy (CRT) when they are selected with standard clinical methods, including those based on QRS duration,1–5 several echocardiographic approaches to the quantification of mechanical dyssynchrony have been tested in an attempt to improve patient selection on the basis of electrocardiographic QRS duration.6 –12 Speckle-tracking echocardiography strain has the advantage of differentiating active contraction from passive motion and is not affected by the Doppler angle of incidence, as in tissue Doppler imaging. Because this technique uses routine grayscale images, it allows the assessment of the 3 different types of speckle-tracking stain (radial, circumferential, and longitudinal). Lim et al13 recently reported the strain delay index determined by longitudinal speckle-tracking strain from standard apical views coupled with the factors of both left
a
Division of Cardiovascular Medicine, Department of Internal Medicine, Kobe University Graduate School of Medicine, Kobe, Japan; and b University of Pittsburgh, Pittsburgh, Pennsylvania. Manuscript received June 4, 2010; revised manuscript received and accepted September 18, 2010. *Corresponding author: Tel: 81-78-382-5846; fax: 81-78-382-5859. E-mail address: [email protected] (H. Tanaka). 0002-9149/11/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.09.038
ventricular (LV) dyssynchrony and residual myocardial contractility. Although this new index was highly predictive of response to CRT, it was calculated from longitudinal strain only and required specific software for analysis. Accordingly, our objective was to test the hypothesis that our relatively simple version of the strain dyssynchrony index (SDI) can predict response to CRT and that a combined assessment of radial, circumferential, and longitudinal SDIs can further enhance the ability of the method to predict response to CRT. Furthermore, we evaluated conventional dyssynchrony measures for a comparison with the ability of the SDI to predict response to CRT. Methods The study included 56 consecutive patients with heart failure in New York Heart Association functional class III or IV with ejection fractions ⱕ35% and QRS durations ⱖ120 ms, who all underwent CRT. Four of these patients were excluded from all subsequent analyses because of inadequate echocardiographic image quality, so that 52 patients were enrolled in this study (Table 1). The mean age was 67 ⫾ 12 years, the mean ejection fraction was 25 ⫾ 7%, and the mean QRS duration was 166 ⫾ 26 ms. Thirteen patients (25%) had ischemic cardiomyopathy, defined as the www.ajconline.org
440
The American Journal of Cardiology (www.ajconline.org)
Table 1 Baseline characteristics of patients and their response to cardiac resynchronization therapy Variable
All Patients (n ⫽ 52)
Responders (n ⫽ 36)
Nonresponders (n ⫽ 16)
p Value
Age (years) Men/women NYHA functional class III/IV QRS duration (ms) SR/AF/paced EDV (ml) ESV (ml) Ejection fraction (%) Heart failure cause Ischemic Nonischemic Medications ACE inhibitors/ARBs  blockers Diuretics Dyssynchrony measurements (ms) Yu index IVMD Radial dyssynchrony by speckle-tracking strain SDI (%) Radial Circumferential Longitudinal
67 ⫾ 12 35/17 44/8 166 ⫾ 26 33/7/12 167 ⫾ 71 128 ⫾ 63 25 ⫾ 7
67 ⫾ 12 25/12 31/5 165 ⫾ 25 22/4/10 164 ⫾ 67 127 ⫾ 62 24 ⫾ 7
65 ⫾ 13 10/6 13/3 168 ⫾ 27 11/3/2 172 ⫾ 82 129 ⫾ 68 27 ⫾ 6
NS NS NS NS NS NS NS NS
13 (25%) 39 (75%)
10 (28%) 26 (72%)
3 (19%) 13 (81%)
NS NS
48 (92%) 44 (85%) 45 (85%)
34 (94%) 30 (88%) 31 (86%)
14 (88%) 14 (88%) 14 (88%)
NS NS NS
47 ⫾ 17 45 ⫾ 25 248 ⫾ 168
48 ⫾ 15 49 ⫾ 25 293 ⫾ 171
45 ⫾ 20 34 ⫾ 21 146 ⫾ 110
NS ⬍0.05 ⬍0.005
8.3 ⫾ 4.6 3.7 ⫾ 1.8 4.0 ⫾ 1.4
10.1 ⫾ 4.4 4.3 ⫾ 1.8 4.4 ⫾ 1.2
4.4 ⫾ 2.3 2.5 ⫾ 1.2 3.0 ⫾ 1.2
⬍0.001 ⬍0.001 ⬍0.001
Data are expressed as mean ⫾ SD or as number (percentage). ACE ⫽ angiotensin-converting enzyme; AF ⫽ atrial fibrillation; ARB ⫽ angiotensin receptor blocker; EDV⫽ end-diastolic volume; IVMD ⫽ interventricular mechanical delay; NYHA ⫽ New York Heart Association; SR ⫽ sinus rhythm.
presence of ⱖ75% stenosis of ⱖ1 major epicardial coronary artery and/or previous coronary revascularization. Thirtythree patients (64%) were diagnosed with sinus rhythm and 5 (10%) with atrial fibrillation. In addition, 14 (24%) had previously undergone the implantation of permanent right ventricular pacing ⱖ1 year before enrollment and features predominantly right ventricular pacing, which was defined as ⱖ90% paced when the device was interrogated at the time of enrollment. All patients were receiving optimal pharmacologic therapy, if tolerated. Written informed consent was obtained from all patients. All echocardiographic studies were performed with a 3.5-MHz transducer using a commercially available echocardiography system (Vivid 7; GE Vingmed Ultrasound AS, Horten, Norway). Patients were studied before and 3 ⫾ 1 months after CRT. Digital routine grayscale 2-dimensional cine loops were obtained, including mid-LV shortaxis views at the level of the papillary muscle, standard apical views (4 chamber, 2 chamber, and long axis), and pulsed Doppler interrogation of the LV outflow tract. Sector width was optimized to allow complete myocardial visualization while maximizing the frame rate. Mean frame rates were 62 ⫾ 7 frames/s in the standard apical view and 65 ⫾ 4 frames/s in the short-axis view for grayscale imaging used for speckle-tracking analysis. For patients with atrial fibrillation, measurements of standard echocardiographic and speckle-tracking parameters were obtained as the averages of ⱖ4 cardiac cycles. Digital data were transferred to dedicated software (EchoPAC version 8.0.0; GE Vingmed Ultrasound AS) for subsequent off-line analysis. LV enddiastolic volume, end-systolic volume (ESV), and ejection
fraction were obtained with the modified biplane Simpson’s method.14 A response to CRT was defined as reverse remodeling detected by a relative decrease of ⱖ15% in ESV at 3-month follow-up. In addition, clinical response was defined as an improvement of ⱖ1 grade in New York Heart Association functional class at 3-month follow-up. Routine pulsed Doppler was used to determine interventricular dyssynchrony, as previously described.15,16 Interventricular mechanical delay was determined as the time difference in the onset of right ventricular ejection velocity to that of LV ejection velocity. An interventricular mechanical delay ⱖ40 ms was considered to constitute significant dyssynchrony.15,16 Longitudinal dyssynchrony was determined using tissue Doppler cine loops from 3 consecutive beats, which were obtained in 3 standard apical views, as previously described in detail.8,11,15 Regions of interest (6 ⫻ 8 mm) were placed on the basal and mid-LV segments for each of the 3 standard views for a 12-site time to peak velocity analysis. Longitudinal tissue Doppler dyssynchrony was determined as the standard deviation of time to peak systolic velocities from the onset of the QRS complex of the 12 sites (Yu index).8,11,15 Longitudinal tissue Doppler dyssynchrony ⱖ32 ms was considered to constitute significant dyssynchrony.8,11,15 For radial dyssynchrony analysis by speckle-tracking strain, routine grayscale mid-LV short-axis images were used, as previously described in detail.9,11 Briefly, an enddiastolic circular region of interest was traced on the endocardial cavity with a point-and-click approach. A second
Heart Failure/Comprehensive Assessment of Strain Dyssynchrony Index
441
(A) Radial Strain Dyssynchrony Index AVC ε-peak (Posterior) ε-peak ( (Anterior Septum)
ε-AVC (Posterior)
ε-AVC (Anterior Septum)
(B) Circumferential Strain Dyssynchrony Index AVC
ε-AVC (Postero Septum)
ε-AVC (Posterior) ε peak ε-peak (Posterior)
ε-peak (Postero Septum)
(C) Longitudinal Strain Dyssynchrony Index AVC
ε-AVC (Mid-Septum)
ε-AVC (Basal-Lateral)
ε-peak (Mid-Septum)
Strain dyssynchrony Index =
ε-peak (Basal-Lateral)
1 n ∑ ( εpeak – εAVC ) n 1
Figure 1. An example of radial (A), circumferential (B), and longitudinal (C) SDIs. The SDI was calculated as the wasted energy per segment due to dyssynchrony, which was determined using the average difference between -peak and -AVC from 6 segments for radial and circumferential SDIs and 18 segments for longitudinal SDI.
larger concentric circle was then automatically generated and manually adjusted near the epicardium. The software automatically divided the mid-LV short-axis and apical images into 6 standard segments and provided the radial, circumferential, and longitudinal speckle-tracking strain curves throughout the cardiac cycle. Radial dyssynchrony by speckle-tracking strain was defined as the time difference between the anteroseptal and posterior wall segmental peak
strains.9 –12,16,17 Radial dyssynchrony ⱖ130 ms was considered to constitute significant dyssynchrony.9 –12,16,17 The SDI represents the average of the wasted energy due to LV dyssynchrony. The concept of our SDI was based on that of the strain delay index as demonstrated by Lim et al.13 Briefly, SDI analysis was performed by speckle-tracking strain using routine grayscale images. Radial and circumferential SDIs were evaluated from 6 standard segments
442
The American Journal of Cardiology (www.ajconline.org)
(A) Radial SDI
% 100 80
80 Se ensitivity
60
60
40
40
20
20
AUC = 0.87 p < 0.001
AUC = 0.81 p < 0.001
0
20 40 60 80 100-Specificity
100 %
0
(D) IVMD
60 40 20
AUC = 0.69 p < 0.05
0 0
20 40 60 80 100-Specificity
100%
0
20 40 60 80 100-Specificity
100%
80 Sensitivity
80 Sensitivity
80 Sensitivity
% 100
AUC = 0.80 p < 0.001
(F) Radial Dyssynchrony % by Speckle Tracking strain 100
(E) Yu Index
% 100
40
0
100%
20 40 60 80 100-Specificity
60
20
0
0
(C) Longitudinal SDI
% 100
80 ensitivity Se
Sensitivity
(B) Circumferential SDI
% 100
60 40 20
AUC = 0.54 p = 0.63 (NS)
0
60 40 20
AUC = 0.74 p < 0.001
0 0
20 40 60 80 100-Specificity
100%
0
20 40 60 80 100-Specificity
100%
Figure 2. Analysis of receiver-operating characteristics curves of radial (A), circumferential (B), and longitudinal (C) SDIs and interventricular mechanical delay (IVMD) (D), Yu index (E), and radial dyssynchrony by speckle-tracking strain (F) for predicting response to CRT, demonstrating that radial SDI was the best predictor of response to CRT. AUC ⫽ area under the curve. Table 2 Association of individual parameters with response to cardiac resynchronization therapy Variable Dyssynchrony measurements IVMD Yu Index Radial dyssynchrony by speckle-tracking strain SDI Radial Circumferential Longitudinal
AUC (95% CI)
Cutoff
Sensitivity (95% CI)
Specificity (95% CI)
0.69 (0.55–0.81) 0.54 (0.40–0.68) 0.74 (0.55–0.85)
40 ms 32 ms 130 ms
67% (49%–81%) 89% (74%–97%) 75% (60%–88%)
56% (30%–80%) 31% (11%–59%) 56% (38%–69%)
0.87 (0.78–0.97) 0.81 (0.68–0.92) 0.80 (0.65–0.88)
6.5% 3.2% 3.6%
81% (74%–100%) 75% (74%–98%) 75% (59%–100%)
81% (70%–97%) 75% (60%–87%) 75% (57%–85%)
AUC ⫽ area under the receiver-operating characteristic curve; CI ⫽ confidence interval. Other abbreviation as in Table 1.
using mid-LV short-axis views. Longitudinal SDI was assessed at basal, mid, and apical levels in the apical 4-chamber, 2-chamber, and long-axis views for a total of 18 sites (Figure 1). The peak of the Q wave on the electrocardiogram was used as the reference time point for end-diastole. The timing of the aortic valve closure (AVC), defined as the end of LV outflow tract detected by pulsed Doppler, was used as the reference time point for end-systole. The wasted energy per segment because of dyssynchrony was expressed as the difference between peak strain (-peak) and AVC strain (-AVC) (Figure 1). This difference increases with greater degrees of dyssynchrony, because the increase of dyssynchrony leads to the decrease in -AVC. The SDI was then calculated as the average of the absolute difference between -peak and -AVC from 6 segments for radial and circumferential SDI and 18 segments for longitudinal SDI. Three consecutive cardiac cycles were recorded and averaged for each measurement. If a segment showed negative radial strain or positive circumferential and longitudinal
strain during the entire cardiac cycle, the difference between -peak and -AVC was assumed to be 0. Biventricular pacing systems were implanted with right ventricular apical leads and LV leads through the coronary sinus in 50 of the patients. The LV leads were placed in the lateral veins of 23 patients, the posterolateral veins of 21 patients, the anterolateral veins of 4 patients, and the middle veins of 2 patients. The epicardial surgical approach was required for the remaining 2 patients. Of the 7 patients with atrial fibrillation, 3 underwent atrioventricular junction ablation at the same time as CRT, and 2 underwent defibrillation first and their atrial fibrillation disappeared after CRT. Device implantation was successful for all patients without any major complications. The atrioventricular interval was adjusted for optimal diastolic filling by Doppler echocardiographic assessment of mitral inflow,18 and the interventricular interval was adjusted by Doppler echocardiographic assessment of LV outflow 8 ⫾ 2 days after implantation.19,20
Heart Failure/Comprehensive Assessment of Strain Dyssynchrony Index
443
Figure 3. Dot plots of percentage reductions in ESV for radial (A), circumferential (B), and longitudinal (C) SDIs, demonstrating significant overall correlations.
The reproducibility of the SDI was tested in 10 randomly selected patients by 2 independent observers for inter- and intraobserver variability. Interobserver variability was assessed by a second observer blinded to the values obtained by the first observer, and intraobserver variability at a different time by an observer blinded to the results of the previous measurements. Inter- and intraobserver variability was expressed as the absolute difference between the measurements divided by their mean value. Continuous variables are expressed as mean ⫾ SD. Group comparisons between before and 3 months after CRT were performed with paired t tests and group comparisons between responders and nonresponders with unpaired Student’s t tests. The diagnostic performance of LV dyssynchrony indexes for predicting response to CRT was evaluated by means of receiver-operating characteristic curve analysis. Univariate linear correlation analysis was used for a comparison of ESV reduction and SDIs. For all tests, p ⬍0.05 was considered statistically significant. All the analyses were performed with commercially available software (SPSS version 15.0; SPSS, Inc., Chicago, Illinois). Results The baseline clinical and echocardiographic characteristics of the 52 patients are listed in Table 1. Response to CRT, defined as a relative decrease in ESV of ⱖ15%, was observed in 36 patients (69%), and the remaining 16 patients (31%) were classified as nonresponders. Compared to nonresponders, responders were more likely to have greater interventricular mechanical delay (49 ⫾ 25 vs 34 ⫾ 21 ms, p ⬍0.05), radial dyssynchrony assessed by speckle-tracking strain (293 ⫾ 171 vs 146 ⫾ 110 ms, p ⬍0.005), and SDI (radial 10.1 ⫾ 4.4% vs 4.4 ⫾ 2.3%, circumferential 4.3 ⫾ 1.8% vs 2.5 ⫾ 1.2%, and longitudinal 4.4 ⫾ 1.2% vs 3.0 ⫾ 1.2%, all p values ⬍0.001; Table 1). In contrast, there was no significant difference in the Yu index between responders and nonresponders (Table 1). Of the individual parameters, including SDI and conventional dyssynchrony measures, radial SDI ⱖ6.5% proved to be the best predictor of response to CRT, with sensitivity of
Table 3 Comparison of strain dyssynchrony index for predicting response to cardiac resynchronization therapy before and after aortic valve closure Variable SDI (including before and after AVC) Radial Circumferential Longitudinal SDI only after AVC Radial Circumferential Longitudinal SDI only before AVC Radial Circumferential Longitudinal
AUC
Cutoff
Sensitivity
Specificity
0.87 0.81 0.80
6.5% 3.2% 3.6%
81% 75% 75%
81% 75% 75%
0.83 0.75 0.75
4.1% 1.8% 2.6%
83% 72% 72%
75% 68% 69%
0.76 0.69* 0.64†
2.7% 1.1% 1.6%
69% 78% 69%
68% 63% 56%
* p ⬍0.05 vs circumferential SDI; † p ⬍0.05 vs longitudinal SDI. Abbreviation as in Table 2.
81%, specificity of 81%, and area under the curve of 0.88 (p ⬍0.001; Figure 2, Table 2). Circumferential SDI ⱖ3.2% and longitudinal SDI ⱖ3.6% were also found to be successful predictors for response to CRT, with sensitivity and specificity of 75% and 75%, for both parameters and areas under the curve of 0.81 and 0.80, respectively (all p values ⬍0.001; Figure 2, Table 2). Although a large scattering of data, particularly for circumferential and longitudinal SDIs, were observed, each SDI correlated significantly with percentage reduction in ESV after CRT (radial r ⫽ 0.64, p ⬍0.001; circumferential r ⫽ 0.40, p ⬍0.005; longitudinal r ⫽ 0.43, p ⬍0.005; Figure 3). In contrast, only the Yu index was not predictive of response to CRT. Using predefined cut-off values of the conventional dyssynchrony parameters (40 ms for interventricular mechanical delay, 32 ms for Yu index, and 130 ms for radial dyssynchrony of speckle-tracking strain), sensitivity for these parameters was 67%, 89%, and 75%, and specificity was 56%, 31%, and 56%, respectively (Table 2). CRT improved each SDI in responders from 10.1 ⫾ 4.4%
444
The American Journal of Cardiology (www.ajconline.org)
and † p<0.001 *vs.p<0.005 Patients with Positivity for Three SDIs
CR RT Resp ponse Rate R (% %)
100 80
100% (19/19) 84% (10/12)
60 40
42% (5/12)
20 0
*
Paents with Posivity for Three SDIs
Paents with Posivity for Two of the Three SDIs
Paents with Posivity for One of the Three SDIs
† 22% (2/9) Paents with Posivity for None of the Three SDIs
Figure 4. Bar graphs showing relation between response rate to CRT and combined assessment of the SDI. Patients with positivity for 3 SDIs (radial ⱖ6.5%, circumferential ⱖ3.2%, and longitudinal ⱖ3.6%) showed 100% CRT response. In contrast, patients with positivity for only 1 or none of the 3 SDIs showed lower CRT response rates of 42% and 22%, respectively (p ⬍0.005 and p ⬍0.001 vs positivity for 3 SDIs).
to 3.9 ⫾ 2.7% for radial SDI, from 4.3 ⫾ 1.8% to 2.6 ⫾ 1.2% for circumferential SDI, and from 4.4 ⫾ 1.2% to 2.6 ⫾ 1.1% in longitudinal SDI (all p values ⬍0.001) but did not improve in nonresponders, from 4.4 ⫾ 2.3% to 3.8 ⫾ 2.0% for radial SDI, from 2.5 ⫾ 1.2% to 2.3 ⫾ 1.5% for circumferential SDI, and from 3.0 ⫾ 1.2% to 3.1 ⫾ 1.3% for longitudinal SDI. Reduction in each SDI with CRT in responders was higher than that in nonresponders (radial 6.2 ⫾ 3.7% vs 0.6 ⫾ 1.2%, circumferential 1.7 ⫾ 1.1% vs 0.2 ⫾ 1.3%, and longitudinal 1.8 ⫾ 1.2% vs 0.0 ⫾ 1.3%, respectively, all p values ⬍0.001). Interestingly, areas under the curve of prediction of response to CRT for each SDI only after AVC were 0.83, 0.75 and 0.75, and those only before AVC were 0.76, 0.69 (p ⬍0.05 vs SDI including both before and after AVC), and 0.64 (p ⬍0.05 vs SDI including both before and after AVC), respectively (Table 3). There were 19 patients (37%) with positive values for all 3 SDIs (radial ⱖ6.5%, circumferential ⱖ3.2%, and longitudinal ⱖ3.6%). This pattern was associated with the highest incidence of LV functional improvement after CRT, with a response rate of 100%. There were 12 patients (23%) with positive values for any 2 of the 3 SDIs, and this pattern was also associated with a high frequency of LV functional improvement after CRT, with a response rate of 84%. In contrast, patients with positivity for only 1 or none of the 3 SDIs were more likely to show a lower frequency of LV functional improvement, with response rates of 42% and 22%, respectively (p ⬍0.005 and p ⬍0.001 vs positivity for all 3 SDIs; Figure 4). We also observed similar results that radial SDI ⱖ6.5% was the best predictor of clinical response to CRT, with sensitivity of 78%, specificity of 80%, and an area under the curve of 0.86 (p ⬍0.001). Similarly, there were 19 patients (37%) with positive values for all 3 SDIs (radial ⱖ6.5%, circumferential ⱖ3.2%, and longitudinal ⱖ3.4%), and this pattern was associated with the
highest incidence of clinical improvement after CRT, with a response rate of 100%. Interobserver variability was 8 ⫾ 4% for radial SDI, 8 ⫾ 7% for circumferential SDI, and 11 ⫾ 9% for longitudinal SDI, and the corresponding values for intraobserver variability were 6 ⫾ 3%, 7 ⫾ 6%, and 8 ⫾ 8%. Statistical analysis confirmed that there were no differences between SDI for inter- and intraobserver variability. Furthermore, inter- and intraobserver variability of ESV were low, at 5 ⫾ 3% and 6 ⫾ 4%, respectively, suggesting that the end point with reduction in ESV in this study was less affected by the variability. Discussion The findings of our study demonstrate that SDI defined by speckle-tracking imaging can successfully predict response to CRT. Of the individual parameters including SDI and conventional dyssynchrony measures, radial SDI proved to be the best predictor of response to CRT. Furthermore, the combined assessment of radial, circumferential, and longitudinal SDI can further enhance the ability of this method to identify patients with a good chance of responding to CRT. CRT is an established therapeutic option for patients with heart failure with severe symptoms and wide QRS complexes.1–5 Despite promising results, roughly 1/3 of patients do not respond to CRT when they are selected with standard clinical methods, including those based on QRS duration. The quantification of LV dyssynchrony has therefore gained prominence as an important potential means to predict patient response. The most widely used technique is LV longitudinal shortening velocities on tissue Doppler imaging from the apical views by echocardiography.6 – 8 However, the ability of echocardiographic measures of dyssynchrony, in particular tissue Doppler imaging, to predict response to CRT has recently been criticized because of the findings by the Predictors of Responders to CRT (PROSPECT) trial.15 Although mechanical dyssynchrony has been associated with response to CRT, other factors, such as LV lead position,9,21–24 scar burden, and myocardial viability,25–28 also influence response to CRT, regardless of dyssynchrony. Previous investigators have reported that patients whose LV lead positions were concordant with the site of the latest mechanical activation as determined by several approaches, including tissue Doppler imaging,23 speckle tracking,9,26 and magnetic resonance imaging,24 showed a better response to CRT than those with discordant LV lead positions. Adelstein et al28 observed that a greater overall scar burden and a larger number of segments with transmural scars caused a diminished response to CRT. Furthermore, we previously found that reduced myocardial systolic function in the region of the LV pacing lead was associated with inadequate LV resynchronization and resulted in nonresponse to CRT.26 Therefore, a comprehensive approach of LV dyssynchrony and myocardial function may well be required to avoid nonresponse to CRT. The strain delay index reported by Lim et al13 was determined by longitudinal speckle-tracking strain from standard apical views. They demonstrated that a longitudinal strain delay index ⱖ25% was strongly predictive of response to CRT, with
Heart Failure/Comprehensive Assessment of Strain Dyssynchrony Index
sensitivity of 95% and specificity of 83%, and correlated with reverse remodeling (ⱖ15% reduction of LV ESV) in ischemic and nonischemic patients. The strain delay index constitutes a marker of LV dyssynchrony and residual myocardial contractility and may help overcome the limitations of several currently used time delay indexes that do not take account of residual myocardial contractility. Myocardial deformation is most commonly referenced in the radial, circumferential, and longitudinal coordinate systems. Because LV dyssynchrony is in fact a 3-dimensional phenomenon, it is thought that these 3 types of deformation do not provide the same information in the failing heart. Because speckle-tracking echocardiography uses routine grayscale images, this technique, unlike tissue Doppler imaging, allows the assessment of 3 different types of stain (radial, circumferential, and longitudinal). Therefore, the utility of radial dyssynchrony by speckle-tracking strain for the quantification of LV dyssynchrony and prediction of response to CRT is well recognized.9 –12,16,17 Our study demonstrated that the combined approach of radial, circumferential, and longitudinal SDIs results in a more complete assessment of SDI than the use of individual SDI parameters or of radial dyssynchrony by speckle-tracking strain alone. In particular, all 3 SDIs positively improved the ability to identify patients with a good chance of responding to CRT. Therefore, a comprehensive approach taking into consideration myocardial deformation including radial, circumferential, and longitudinal function may prove to be even more useful for enhancement of the ability to identify patients with a good chance of responding to CRT. The main difference between our SDI and Lim et al’s13 strain delay index is that the strain delay index requires specific software for analysis. Our index, the SDI, can be calculated as the average of the wasted energy per segment due to dyssynchrony, which can be determined by the difference between -peak and -AVC, without the need for specific software, and can be applied to other manufacturers’ speckle-tracking systems. Another difference between these 2 methods is the definition of end-systole. They used the timing of the peak global strain to determine the timing of end-systole, but the timing of end-systole could be difference in a patient by radial, circumferential, and longitudinal strain curve. We determined the timing of the AVC using pulsed Doppler wave of LV outflow as end-systole, and it was recognized as “physiologic end-systole,” including whole cardiac function. In addition, our SDI expressed average of wasted energy, which does not depend on the number of segments. Finally, they assumed to be 0 for a segment that exhibited biphasic longitudinal strain with a peak positive value greater than a maximal absolute negative value. However, we did not assume to be 0 for these segments, because we recognized that the segment with biphasic strain curve was not only dyskinesia but also severe early or delayed contraction, which was the 1 of the important factor of response to CRT. In this study, the prediction of response to CRT in SDI only after AVC was higher than that in SDI only before AVC, but lower than that in our original SDI including both before and after AVC. These findings indicated that the wasted energy after AVC might be important for CRT than that before AVC, because any contraction after AVC did not
445
contribute to cardiac output. Meanwhile, the total wasted energy both before and after AVC might be the most important for CRT. This study covered a small number of patients at a single center, so future studies of larger patient populations are necessary to determine the utility of the SDI for predicting response to CRT. All echocardiographic examinations must be evaluated by an observer blinded to the patients’ characteristics. We also observed a similar utility of radial SDI in patients with wider QRS duration (⬎150 ms; n ⫽ 39), which predicted response to CRT with sensitivity of 82%, specificity of 82%, and an area under the curve of 0.88. In contrast, the utility of SDI in patients with narrower QRS duration (⬍130 ms) was unknown, because we included only 2 patients with QRS durations of 120 to 130 ms in this study. The precise reason for worse sensitivity and specificity than strain delay index observed using our SDI is unknown but may include differences in the general population of the study. Moreover, the strain delay index may be computed automatically with less reproducibility. Limitations of the speckle-tracking method include endocardial border tracing, which requires care when manually finetuning the regions of interest to capture the early septal motion followed by later posterior or lateral wall motion in patients with heart failure with wide QRS complexes and adjustment of its width for dyssynchrony analysis before generating and measuring regional strain. Furthermore, speckle-tracking echocardiography requires experience to achieve sufficiently reproducible results and requires raining.15,29 Another limitation is that only 25% of patients in this study had ischemic cardiomyopathy. A possible reason might be the difference in cause between patients with chronic severe heart failure in Japan and other countries, especially the United States and Europe. Osada et al30 reported that 80% of patients with chronic severe heart failure among 284 applicants for heart transplantation in Japan were diagnosed with idiopathic dilated cardiomyopathy. Acknowledgment: We are grateful for the support of the entire staff of the echocardiography and electrophysiology laboratories of Kobe University Hospital. 1. Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E, Kocovic DZ, Packer M, Clavell AL, Hayes DL, Ellestad M, Trupp RJ, Underwood J, Pickering F, Truex C, McAtee P, Messenger J. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002;346: 1845–1853. 2. Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco T, Carson P, DiCarlo L, DeMets D, White BG, DeVries DW, Feldman AM. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004;350:2140 –2150. 3. Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L, Tavazzi L. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005;352:1539 – 1549. 4. Linde C, Leclercq C, Rex S, Garrigue S, Lavergne T, Cazeau S, McKenna W, Fitzgerald M, Deharo JC, Alonso C, Walker S, Braunschweig F, Bailleul C, Daubert JC. Long-term benefits of biventricular pacing in congestive heart failure: results from the Multisite Stimulation in Cardiomyopathy (MUSTIC) study. J Am Coll Cardiol 2002; 40:111–118.
446
The American Journal of Cardiology (www.ajconline.org)
5. Epstein AE, DiMarco JP, Ellenbogen KA, Estes NA III, Freedman RA, Gettes LS, Gillinov AM, Gregoratos G, Hammill SC, Hayes DL, Hlatky MA, Newby LK, Page RL, Schoenfeld MH, Silka MJ, Stevenson LW, Sweeney MO, Smith SC Jr, Jacobs AK, Adams CD, Anderson JL, Buller CE, Creager MA, Ettinger SM, Faxon DP, Halperin JL, Hiratzka LF, Hunt SA, Krumholz HM, Kushner FG, Lytle BW, Nishimura RA, Ornato JP, Riegel B, Tarkington LG, Yancy CW. ACC/ AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the ACC/AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and Antiarrhythmia Devices) developed in collaboration with the American Association for Thoracic Surgery and Society of Thoracic Surgeons. J Am Coll Cardiol 2008; 51:e1– e62. 6. Bax JJ, Bleeker GB, Marwick TH, Molhoek SG, Boersma E, Steendijk P, van der Wall EE, Schalij MJ. Left ventricular dyssynchrony predicts response and prognosis after cardiac resynchronization therapy. J Am Coll Cardiol 2004;44:1834 –1840. 7. Gorcsan J III, Kanzaki H, Bazaz R, Dohi K, Schwartzman D. Usefulness of echocardiographic tissue synchronization imaging to predict acute response to cardiac resynchronization therapy. Am J Cardiol 2004;93:1178 –1181. 8. Yu CM, Chau E, Sanderson JE, Fan K, Tang MO, Fung WH, Lin H, Kong SL, Lam YM, Hill MR, Lau CP. Tissue Doppler echocardiographic evidence of reverse remodeling and improved synchronicity by simultaneously delaying regional contraction after biventricular pacing therapy in heart failure. Circulation 2002;105:438 – 445. 9. Suffoletto MS, Dohi K, Cannesson M, Saba S, Gorcsan J III. Novel speckle-tracking radial strain from routine black-and-white echocardiographic images to quantify dyssynchrony and predict response to cardiac resynchronization therapy. Circulation 2006;113:960 –968. 10. Tanaka H, Hara H, Saba S, Gorcsan J III. Prediction of response to cardiac resynchronization therapy by speckle tracking echocardiography using different software approaches. J Am Soc Echocardiogr 2009;22:677– 684. 11. Gorcsan J III, Abraham T, Agler DA, Bax JJ, Derumeaux G, Grimm RA, Martin R, Steinberg JS, Sutton MS, Yu CM. Echocardiography for cardiac resynchronization therapy: recommendations for performance and reporting—a report from the American Society of Echocardiography Dyssynchrony Writing Group endorsed by the Heart Rhythm Society. J Am Soc Echocardiogr 2008;21:191–213. 12. Delgado V, Ypenburg C, van Bommel RJ, Tops LF, Mollema SA, Marsan NA, Bleeker GB, Schalij MJ, Bax JJ. Assessment of left ventricular dyssynchrony by speckle tracking strain imaging comparison between longitudinal, circumferential, and radial strain in cardiac resynchronization therapy. J Am Coll Cardiol 2008;51:1944 –1952. 13. Lim P, Buakhamsri A, Popovic ZB, Greenberg NL, Patel D, Thomas JD, Grimm RA. Longitudinal strain delay index by speckle tracking imaging: a new marker of response to cardiac resynchronization therapy. Circulation 2008;118:1130 –1137. 14. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA, Picard MH, Roman MJ, Seward J, Shanewise JS, Solomon SD, Spencer KT, Sutton MS, Stewart WJ. Recommendations for chamber quantification: a report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005;18:1440 –1463. 15. Chung ES, Leon AR, Tavazzi L, Sun JP, Nihoyannopoulos P, Merlino J, Abraham WT, Ghio S, Leclercq C, Bax JJ, Yu CM, Gorcsan J III,
16.
17.
18.
19. 20.
21.
22.
23.
24.
25.
26.
27.
28. 29. 30.
St John Sutton M, De Sutter J, Murillo J. Results of the Predictors of Response to CRT (PROSPECT) trial. Circulation 2008;117: 2608 –2616. Gorcsan J III, Tanabe M, Bleeker GB, Suffoletto MS, Thomas NC, Saba S, Tops LF, Schalij MJ, Bax JJ. Combined longitudinal and radial dyssynchrony predicts ventricular response after resynchronization therapy. J Am Coll Cardiol 2007;50:1476 –1483. Tops LF, Suffoletto MS, Bleeker GB, Boersma E, van der Wall EE, Gorcsan J III, Schalij MJ, Bax JJ. Speckle-tracking radial strain reveals left ventricular dyssynchrony in patients with permanent right ventricular pacing. J Am Coll Cardiol 2007;50:1180 –1188. Ishikawa T, Sumita S, Kimura K, Kikuchi M, Kosuge M, Kuji N, Endo T, Sugano T, Sigemasa T, Kobayashi I, Tochikubo O, Usui T. Prediction of optimal atrioventricular delay in patients with implanted DDD pacemakers. Pacing Clin Electrophysiol 1999;22:1365–1371. van Gelder BM, Bracke FA, Meijer A, Lakerveld LJ, Pijls NH. Effect of optimizing the VV interval on left ventricular contractility in cardiac resynchronization therapy. Am J Cardiol 2004;93:1500 –1503. Perego GB, Chianca R, Facchini M, Frattola A, Balla E, Zucchi S, Cavaglia S, Vicini I, Negretto M, Osculati G. Simultaneous vs. sequential biventricular pacing in dilated cardiomyopathy: an acute hemodynamic study. Eur J Heart Fail 2003;5:305–313. Becker M, Kramann R, Franke A, Breithardt OA, Heussen N, Knackstedt C, Stellbrink C, Schauerte P, Kelm M, Hoffmann R. Impact of left ventricular lead position in cardiac resynchronization therapy on left ventricular remodelling. A circumferential strain analysis based on 2D echocardiography. Eur Heart J 2007;28:1211–1220. Song ZZ. Impact of left ventricular lead position in cardiac resynchronization therapy on left ventricular remodelling: a circumferential strain analysis based on 2D echocardiography. Eur Heart J 2008;29: 684. Ansalone G, Giannantoni P, Ricci R, Trambaiolo P, Fedele F, Santini M. Doppler myocardial imaging to evaluate the effectiveness of pacing sites in patients receiving biventricular pacing. J Am Coll Cardiol 2002;39:489 – 499. Helm RH, Byrne M, Helm PA, Daya SK, Osman NF, Tunin R, Halperin HR, Berger RD, Kass DA, Lardo AC. Three-dimensional mapping of optimal left ventricular pacing site for cardiac resynchronization. Circulation 2007;115:953–961. Bleeker GB, Kaandorp TA, Lamb HJ, Boersma E, Steendijk P, de Roos A, van der Wall EE, Schalij MJ, Bax JJ. Effect of posterolateral scar tissue on clinical and echocardiographic improvement after cardiac resynchronization therapy. Circulation 2006;113:969 –976. Norisada K, Kawai H, Tanaka H, Tatsumi K, Onishi T, Fukuzawa K, Yoshida A, Hirata KI. Myocardial Contractile Function in the Region of the Left Ventricular Pacing Lead Predicts the Response to Cardiac Resynchronization Therapy Assessed by Two-Dimensional Speckle Tracking Echocardiography. J Am Soc Echocardiogr 2010;23:181– 189. Ypenburg C, Schalij MJ, Bleeker GB, Steendijk P, Boersma E, DibbetsSchneider P, Stokkel MP, van der Wall EE, Bax JJ. Impact of viability and scar tissue on response to cardiac resynchronization therapy in ischaemic heart failure patients. Eur Heart J 2007;28:33– 41. Adelstein EC, Saba S. Scar burden by myocardial perfusion imaging predicts echocardiographic response to cardiac resynchronization therapy in ischemic cardiomyopathy. Am Heart J 2007;153:105–112. Yu CM, Bax JJ, Gorcsan J III. Critical appraisal of methods to assess mechanical dyssynchrony. Curr Opin Cardiol 2009;24:18 –28. Osada K, Imaizumi T. Special report from the heart transplant candidate registry committee in Japan. J Heart Lung Transplant 2005;24: 810 – 814.
a
Division of Cardiovascular Medicine, Department of Internal Medicine, Kobe University Graduate School of Medicine, Kobe, Japan; and b University of Pittsburgh, Pittsburgh, Pennsylvania. Manuscript received June 4, 2010; revised manuscript received and accepted September 18, 2010. *Corresponding author: Tel: 81-78-382-5846; fax: 81-78-382-5859. E-mail address: [email protected] (H. Tanaka). 0002-9149/11/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.09.038
ventricular (LV) dyssynchrony and residual myocardial contractility. Although this new index was highly predictive of response to CRT, it was calculated from longitudinal strain only and required specific software for analysis. Accordingly, our objective was to test the hypothesis that our relatively simple version of the strain dyssynchrony index (SDI) can predict response to CRT and that a combined assessment of radial, circumferential, and longitudinal SDIs can further enhance the ability of the method to predict response to CRT. Furthermore, we evaluated conventional dyssynchrony measures for a comparison with the ability of the SDI to predict response to CRT. Methods The study included 56 consecutive patients with heart failure in New York Heart Association functional class III or IV with ejection fractions ⱕ35% and QRS durations ⱖ120 ms, who all underwent CRT. Four of these patients were excluded from all subsequent analyses because of inadequate echocardiographic image quality, so that 52 patients were enrolled in this study (Table 1). The mean age was 67 ⫾ 12 years, the mean ejection fraction was 25 ⫾ 7%, and the mean QRS duration was 166 ⫾ 26 ms. Thirteen patients (25%) had ischemic cardiomyopathy, defined as the www.ajconline.org
440
The American Journal of Cardiology (www.ajconline.org)
Table 1 Baseline characteristics of patients and their response to cardiac resynchronization therapy Variable
All Patients (n ⫽ 52)
Responders (n ⫽ 36)
Nonresponders (n ⫽ 16)
p Value
Age (years) Men/women NYHA functional class III/IV QRS duration (ms) SR/AF/paced EDV (ml) ESV (ml) Ejection fraction (%) Heart failure cause Ischemic Nonischemic Medications ACE inhibitors/ARBs  blockers Diuretics Dyssynchrony measurements (ms) Yu index IVMD Radial dyssynchrony by speckle-tracking strain SDI (%) Radial Circumferential Longitudinal
67 ⫾ 12 35/17 44/8 166 ⫾ 26 33/7/12 167 ⫾ 71 128 ⫾ 63 25 ⫾ 7
67 ⫾ 12 25/12 31/5 165 ⫾ 25 22/4/10 164 ⫾ 67 127 ⫾ 62 24 ⫾ 7
65 ⫾ 13 10/6 13/3 168 ⫾ 27 11/3/2 172 ⫾ 82 129 ⫾ 68 27 ⫾ 6
NS NS NS NS NS NS NS NS
13 (25%) 39 (75%)
10 (28%) 26 (72%)
3 (19%) 13 (81%)
NS NS
48 (92%) 44 (85%) 45 (85%)
34 (94%) 30 (88%) 31 (86%)
14 (88%) 14 (88%) 14 (88%)
NS NS NS
47 ⫾ 17 45 ⫾ 25 248 ⫾ 168
48 ⫾ 15 49 ⫾ 25 293 ⫾ 171
45 ⫾ 20 34 ⫾ 21 146 ⫾ 110
NS ⬍0.05 ⬍0.005
8.3 ⫾ 4.6 3.7 ⫾ 1.8 4.0 ⫾ 1.4
10.1 ⫾ 4.4 4.3 ⫾ 1.8 4.4 ⫾ 1.2
4.4 ⫾ 2.3 2.5 ⫾ 1.2 3.0 ⫾ 1.2
⬍0.001 ⬍0.001 ⬍0.001
Data are expressed as mean ⫾ SD or as number (percentage). ACE ⫽ angiotensin-converting enzyme; AF ⫽ atrial fibrillation; ARB ⫽ angiotensin receptor blocker; EDV⫽ end-diastolic volume; IVMD ⫽ interventricular mechanical delay; NYHA ⫽ New York Heart Association; SR ⫽ sinus rhythm.
presence of ⱖ75% stenosis of ⱖ1 major epicardial coronary artery and/or previous coronary revascularization. Thirtythree patients (64%) were diagnosed with sinus rhythm and 5 (10%) with atrial fibrillation. In addition, 14 (24%) had previously undergone the implantation of permanent right ventricular pacing ⱖ1 year before enrollment and features predominantly right ventricular pacing, which was defined as ⱖ90% paced when the device was interrogated at the time of enrollment. All patients were receiving optimal pharmacologic therapy, if tolerated. Written informed consent was obtained from all patients. All echocardiographic studies were performed with a 3.5-MHz transducer using a commercially available echocardiography system (Vivid 7; GE Vingmed Ultrasound AS, Horten, Norway). Patients were studied before and 3 ⫾ 1 months after CRT. Digital routine grayscale 2-dimensional cine loops were obtained, including mid-LV shortaxis views at the level of the papillary muscle, standard apical views (4 chamber, 2 chamber, and long axis), and pulsed Doppler interrogation of the LV outflow tract. Sector width was optimized to allow complete myocardial visualization while maximizing the frame rate. Mean frame rates were 62 ⫾ 7 frames/s in the standard apical view and 65 ⫾ 4 frames/s in the short-axis view for grayscale imaging used for speckle-tracking analysis. For patients with atrial fibrillation, measurements of standard echocardiographic and speckle-tracking parameters were obtained as the averages of ⱖ4 cardiac cycles. Digital data were transferred to dedicated software (EchoPAC version 8.0.0; GE Vingmed Ultrasound AS) for subsequent off-line analysis. LV enddiastolic volume, end-systolic volume (ESV), and ejection
fraction were obtained with the modified biplane Simpson’s method.14 A response to CRT was defined as reverse remodeling detected by a relative decrease of ⱖ15% in ESV at 3-month follow-up. In addition, clinical response was defined as an improvement of ⱖ1 grade in New York Heart Association functional class at 3-month follow-up. Routine pulsed Doppler was used to determine interventricular dyssynchrony, as previously described.15,16 Interventricular mechanical delay was determined as the time difference in the onset of right ventricular ejection velocity to that of LV ejection velocity. An interventricular mechanical delay ⱖ40 ms was considered to constitute significant dyssynchrony.15,16 Longitudinal dyssynchrony was determined using tissue Doppler cine loops from 3 consecutive beats, which were obtained in 3 standard apical views, as previously described in detail.8,11,15 Regions of interest (6 ⫻ 8 mm) were placed on the basal and mid-LV segments for each of the 3 standard views for a 12-site time to peak velocity analysis. Longitudinal tissue Doppler dyssynchrony was determined as the standard deviation of time to peak systolic velocities from the onset of the QRS complex of the 12 sites (Yu index).8,11,15 Longitudinal tissue Doppler dyssynchrony ⱖ32 ms was considered to constitute significant dyssynchrony.8,11,15 For radial dyssynchrony analysis by speckle-tracking strain, routine grayscale mid-LV short-axis images were used, as previously described in detail.9,11 Briefly, an enddiastolic circular region of interest was traced on the endocardial cavity with a point-and-click approach. A second
Heart Failure/Comprehensive Assessment of Strain Dyssynchrony Index
441
(A) Radial Strain Dyssynchrony Index AVC ε-peak (Posterior) ε-peak ( (Anterior Septum)
ε-AVC (Posterior)
ε-AVC (Anterior Septum)
(B) Circumferential Strain Dyssynchrony Index AVC
ε-AVC (Postero Septum)
ε-AVC (Posterior) ε peak ε-peak (Posterior)
ε-peak (Postero Septum)
(C) Longitudinal Strain Dyssynchrony Index AVC
ε-AVC (Mid-Septum)
ε-AVC (Basal-Lateral)
ε-peak (Mid-Septum)
Strain dyssynchrony Index =
ε-peak (Basal-Lateral)
1 n ∑ ( εpeak – εAVC ) n 1
Figure 1. An example of radial (A), circumferential (B), and longitudinal (C) SDIs. The SDI was calculated as the wasted energy per segment due to dyssynchrony, which was determined using the average difference between -peak and -AVC from 6 segments for radial and circumferential SDIs and 18 segments for longitudinal SDI.
larger concentric circle was then automatically generated and manually adjusted near the epicardium. The software automatically divided the mid-LV short-axis and apical images into 6 standard segments and provided the radial, circumferential, and longitudinal speckle-tracking strain curves throughout the cardiac cycle. Radial dyssynchrony by speckle-tracking strain was defined as the time difference between the anteroseptal and posterior wall segmental peak
strains.9 –12,16,17 Radial dyssynchrony ⱖ130 ms was considered to constitute significant dyssynchrony.9 –12,16,17 The SDI represents the average of the wasted energy due to LV dyssynchrony. The concept of our SDI was based on that of the strain delay index as demonstrated by Lim et al.13 Briefly, SDI analysis was performed by speckle-tracking strain using routine grayscale images. Radial and circumferential SDIs were evaluated from 6 standard segments
442
The American Journal of Cardiology (www.ajconline.org)
(A) Radial SDI
% 100 80
80 Se ensitivity
60
60
40
40
20
20
AUC = 0.87 p < 0.001
AUC = 0.81 p < 0.001
0
20 40 60 80 100-Specificity
100 %
0
(D) IVMD
60 40 20
AUC = 0.69 p < 0.05
0 0
20 40 60 80 100-Specificity
100%
0
20 40 60 80 100-Specificity
100%
80 Sensitivity
80 Sensitivity
80 Sensitivity
% 100
AUC = 0.80 p < 0.001
(F) Radial Dyssynchrony % by Speckle Tracking strain 100
(E) Yu Index
% 100
40
0
100%
20 40 60 80 100-Specificity
60
20
0
0
(C) Longitudinal SDI
% 100
80 ensitivity Se
Sensitivity
(B) Circumferential SDI
% 100
60 40 20
AUC = 0.54 p = 0.63 (NS)
0
60 40 20
AUC = 0.74 p < 0.001
0 0
20 40 60 80 100-Specificity
100%
0
20 40 60 80 100-Specificity
100%
Figure 2. Analysis of receiver-operating characteristics curves of radial (A), circumferential (B), and longitudinal (C) SDIs and interventricular mechanical delay (IVMD) (D), Yu index (E), and radial dyssynchrony by speckle-tracking strain (F) for predicting response to CRT, demonstrating that radial SDI was the best predictor of response to CRT. AUC ⫽ area under the curve. Table 2 Association of individual parameters with response to cardiac resynchronization therapy Variable Dyssynchrony measurements IVMD Yu Index Radial dyssynchrony by speckle-tracking strain SDI Radial Circumferential Longitudinal
AUC (95% CI)
Cutoff
Sensitivity (95% CI)
Specificity (95% CI)
0.69 (0.55–0.81) 0.54 (0.40–0.68) 0.74 (0.55–0.85)
40 ms 32 ms 130 ms
67% (49%–81%) 89% (74%–97%) 75% (60%–88%)
56% (30%–80%) 31% (11%–59%) 56% (38%–69%)
0.87 (0.78–0.97) 0.81 (0.68–0.92) 0.80 (0.65–0.88)
6.5% 3.2% 3.6%
81% (74%–100%) 75% (74%–98%) 75% (59%–100%)
81% (70%–97%) 75% (60%–87%) 75% (57%–85%)
AUC ⫽ area under the receiver-operating characteristic curve; CI ⫽ confidence interval. Other abbreviation as in Table 1.
using mid-LV short-axis views. Longitudinal SDI was assessed at basal, mid, and apical levels in the apical 4-chamber, 2-chamber, and long-axis views for a total of 18 sites (Figure 1). The peak of the Q wave on the electrocardiogram was used as the reference time point for end-diastole. The timing of the aortic valve closure (AVC), defined as the end of LV outflow tract detected by pulsed Doppler, was used as the reference time point for end-systole. The wasted energy per segment because of dyssynchrony was expressed as the difference between peak strain (-peak) and AVC strain (-AVC) (Figure 1). This difference increases with greater degrees of dyssynchrony, because the increase of dyssynchrony leads to the decrease in -AVC. The SDI was then calculated as the average of the absolute difference between -peak and -AVC from 6 segments for radial and circumferential SDI and 18 segments for longitudinal SDI. Three consecutive cardiac cycles were recorded and averaged for each measurement. If a segment showed negative radial strain or positive circumferential and longitudinal
strain during the entire cardiac cycle, the difference between -peak and -AVC was assumed to be 0. Biventricular pacing systems were implanted with right ventricular apical leads and LV leads through the coronary sinus in 50 of the patients. The LV leads were placed in the lateral veins of 23 patients, the posterolateral veins of 21 patients, the anterolateral veins of 4 patients, and the middle veins of 2 patients. The epicardial surgical approach was required for the remaining 2 patients. Of the 7 patients with atrial fibrillation, 3 underwent atrioventricular junction ablation at the same time as CRT, and 2 underwent defibrillation first and their atrial fibrillation disappeared after CRT. Device implantation was successful for all patients without any major complications. The atrioventricular interval was adjusted for optimal diastolic filling by Doppler echocardiographic assessment of mitral inflow,18 and the interventricular interval was adjusted by Doppler echocardiographic assessment of LV outflow 8 ⫾ 2 days after implantation.19,20
Heart Failure/Comprehensive Assessment of Strain Dyssynchrony Index
443
Figure 3. Dot plots of percentage reductions in ESV for radial (A), circumferential (B), and longitudinal (C) SDIs, demonstrating significant overall correlations.
The reproducibility of the SDI was tested in 10 randomly selected patients by 2 independent observers for inter- and intraobserver variability. Interobserver variability was assessed by a second observer blinded to the values obtained by the first observer, and intraobserver variability at a different time by an observer blinded to the results of the previous measurements. Inter- and intraobserver variability was expressed as the absolute difference between the measurements divided by their mean value. Continuous variables are expressed as mean ⫾ SD. Group comparisons between before and 3 months after CRT were performed with paired t tests and group comparisons between responders and nonresponders with unpaired Student’s t tests. The diagnostic performance of LV dyssynchrony indexes for predicting response to CRT was evaluated by means of receiver-operating characteristic curve analysis. Univariate linear correlation analysis was used for a comparison of ESV reduction and SDIs. For all tests, p ⬍0.05 was considered statistically significant. All the analyses were performed with commercially available software (SPSS version 15.0; SPSS, Inc., Chicago, Illinois). Results The baseline clinical and echocardiographic characteristics of the 52 patients are listed in Table 1. Response to CRT, defined as a relative decrease in ESV of ⱖ15%, was observed in 36 patients (69%), and the remaining 16 patients (31%) were classified as nonresponders. Compared to nonresponders, responders were more likely to have greater interventricular mechanical delay (49 ⫾ 25 vs 34 ⫾ 21 ms, p ⬍0.05), radial dyssynchrony assessed by speckle-tracking strain (293 ⫾ 171 vs 146 ⫾ 110 ms, p ⬍0.005), and SDI (radial 10.1 ⫾ 4.4% vs 4.4 ⫾ 2.3%, circumferential 4.3 ⫾ 1.8% vs 2.5 ⫾ 1.2%, and longitudinal 4.4 ⫾ 1.2% vs 3.0 ⫾ 1.2%, all p values ⬍0.001; Table 1). In contrast, there was no significant difference in the Yu index between responders and nonresponders (Table 1). Of the individual parameters, including SDI and conventional dyssynchrony measures, radial SDI ⱖ6.5% proved to be the best predictor of response to CRT, with sensitivity of
Table 3 Comparison of strain dyssynchrony index for predicting response to cardiac resynchronization therapy before and after aortic valve closure Variable SDI (including before and after AVC) Radial Circumferential Longitudinal SDI only after AVC Radial Circumferential Longitudinal SDI only before AVC Radial Circumferential Longitudinal
AUC
Cutoff
Sensitivity
Specificity
0.87 0.81 0.80
6.5% 3.2% 3.6%
81% 75% 75%
81% 75% 75%
0.83 0.75 0.75
4.1% 1.8% 2.6%
83% 72% 72%
75% 68% 69%
0.76 0.69* 0.64†
2.7% 1.1% 1.6%
69% 78% 69%
68% 63% 56%
* p ⬍0.05 vs circumferential SDI; † p ⬍0.05 vs longitudinal SDI. Abbreviation as in Table 2.
81%, specificity of 81%, and area under the curve of 0.88 (p ⬍0.001; Figure 2, Table 2). Circumferential SDI ⱖ3.2% and longitudinal SDI ⱖ3.6% were also found to be successful predictors for response to CRT, with sensitivity and specificity of 75% and 75%, for both parameters and areas under the curve of 0.81 and 0.80, respectively (all p values ⬍0.001; Figure 2, Table 2). Although a large scattering of data, particularly for circumferential and longitudinal SDIs, were observed, each SDI correlated significantly with percentage reduction in ESV after CRT (radial r ⫽ 0.64, p ⬍0.001; circumferential r ⫽ 0.40, p ⬍0.005; longitudinal r ⫽ 0.43, p ⬍0.005; Figure 3). In contrast, only the Yu index was not predictive of response to CRT. Using predefined cut-off values of the conventional dyssynchrony parameters (40 ms for interventricular mechanical delay, 32 ms for Yu index, and 130 ms for radial dyssynchrony of speckle-tracking strain), sensitivity for these parameters was 67%, 89%, and 75%, and specificity was 56%, 31%, and 56%, respectively (Table 2). CRT improved each SDI in responders from 10.1 ⫾ 4.4%
444
The American Journal of Cardiology (www.ajconline.org)
and † p<0.001 *vs.p<0.005 Patients with Positivity for Three SDIs
CR RT Resp ponse Rate R (% %)
100 80
100% (19/19) 84% (10/12)
60 40
42% (5/12)
20 0
*
Paents with Posivity for Three SDIs
Paents with Posivity for Two of the Three SDIs
Paents with Posivity for One of the Three SDIs
† 22% (2/9) Paents with Posivity for None of the Three SDIs
Figure 4. Bar graphs showing relation between response rate to CRT and combined assessment of the SDI. Patients with positivity for 3 SDIs (radial ⱖ6.5%, circumferential ⱖ3.2%, and longitudinal ⱖ3.6%) showed 100% CRT response. In contrast, patients with positivity for only 1 or none of the 3 SDIs showed lower CRT response rates of 42% and 22%, respectively (p ⬍0.005 and p ⬍0.001 vs positivity for 3 SDIs).
to 3.9 ⫾ 2.7% for radial SDI, from 4.3 ⫾ 1.8% to 2.6 ⫾ 1.2% for circumferential SDI, and from 4.4 ⫾ 1.2% to 2.6 ⫾ 1.1% in longitudinal SDI (all p values ⬍0.001) but did not improve in nonresponders, from 4.4 ⫾ 2.3% to 3.8 ⫾ 2.0% for radial SDI, from 2.5 ⫾ 1.2% to 2.3 ⫾ 1.5% for circumferential SDI, and from 3.0 ⫾ 1.2% to 3.1 ⫾ 1.3% for longitudinal SDI. Reduction in each SDI with CRT in responders was higher than that in nonresponders (radial 6.2 ⫾ 3.7% vs 0.6 ⫾ 1.2%, circumferential 1.7 ⫾ 1.1% vs 0.2 ⫾ 1.3%, and longitudinal 1.8 ⫾ 1.2% vs 0.0 ⫾ 1.3%, respectively, all p values ⬍0.001). Interestingly, areas under the curve of prediction of response to CRT for each SDI only after AVC were 0.83, 0.75 and 0.75, and those only before AVC were 0.76, 0.69 (p ⬍0.05 vs SDI including both before and after AVC), and 0.64 (p ⬍0.05 vs SDI including both before and after AVC), respectively (Table 3). There were 19 patients (37%) with positive values for all 3 SDIs (radial ⱖ6.5%, circumferential ⱖ3.2%, and longitudinal ⱖ3.6%). This pattern was associated with the highest incidence of LV functional improvement after CRT, with a response rate of 100%. There were 12 patients (23%) with positive values for any 2 of the 3 SDIs, and this pattern was also associated with a high frequency of LV functional improvement after CRT, with a response rate of 84%. In contrast, patients with positivity for only 1 or none of the 3 SDIs were more likely to show a lower frequency of LV functional improvement, with response rates of 42% and 22%, respectively (p ⬍0.005 and p ⬍0.001 vs positivity for all 3 SDIs; Figure 4). We also observed similar results that radial SDI ⱖ6.5% was the best predictor of clinical response to CRT, with sensitivity of 78%, specificity of 80%, and an area under the curve of 0.86 (p ⬍0.001). Similarly, there were 19 patients (37%) with positive values for all 3 SDIs (radial ⱖ6.5%, circumferential ⱖ3.2%, and longitudinal ⱖ3.4%), and this pattern was associated with the
highest incidence of clinical improvement after CRT, with a response rate of 100%. Interobserver variability was 8 ⫾ 4% for radial SDI, 8 ⫾ 7% for circumferential SDI, and 11 ⫾ 9% for longitudinal SDI, and the corresponding values for intraobserver variability were 6 ⫾ 3%, 7 ⫾ 6%, and 8 ⫾ 8%. Statistical analysis confirmed that there were no differences between SDI for inter- and intraobserver variability. Furthermore, inter- and intraobserver variability of ESV were low, at 5 ⫾ 3% and 6 ⫾ 4%, respectively, suggesting that the end point with reduction in ESV in this study was less affected by the variability. Discussion The findings of our study demonstrate that SDI defined by speckle-tracking imaging can successfully predict response to CRT. Of the individual parameters including SDI and conventional dyssynchrony measures, radial SDI proved to be the best predictor of response to CRT. Furthermore, the combined assessment of radial, circumferential, and longitudinal SDI can further enhance the ability of this method to identify patients with a good chance of responding to CRT. CRT is an established therapeutic option for patients with heart failure with severe symptoms and wide QRS complexes.1–5 Despite promising results, roughly 1/3 of patients do not respond to CRT when they are selected with standard clinical methods, including those based on QRS duration. The quantification of LV dyssynchrony has therefore gained prominence as an important potential means to predict patient response. The most widely used technique is LV longitudinal shortening velocities on tissue Doppler imaging from the apical views by echocardiography.6 – 8 However, the ability of echocardiographic measures of dyssynchrony, in particular tissue Doppler imaging, to predict response to CRT has recently been criticized because of the findings by the Predictors of Responders to CRT (PROSPECT) trial.15 Although mechanical dyssynchrony has been associated with response to CRT, other factors, such as LV lead position,9,21–24 scar burden, and myocardial viability,25–28 also influence response to CRT, regardless of dyssynchrony. Previous investigators have reported that patients whose LV lead positions were concordant with the site of the latest mechanical activation as determined by several approaches, including tissue Doppler imaging,23 speckle tracking,9,26 and magnetic resonance imaging,24 showed a better response to CRT than those with discordant LV lead positions. Adelstein et al28 observed that a greater overall scar burden and a larger number of segments with transmural scars caused a diminished response to CRT. Furthermore, we previously found that reduced myocardial systolic function in the region of the LV pacing lead was associated with inadequate LV resynchronization and resulted in nonresponse to CRT.26 Therefore, a comprehensive approach of LV dyssynchrony and myocardial function may well be required to avoid nonresponse to CRT. The strain delay index reported by Lim et al13 was determined by longitudinal speckle-tracking strain from standard apical views. They demonstrated that a longitudinal strain delay index ⱖ25% was strongly predictive of response to CRT, with
Heart Failure/Comprehensive Assessment of Strain Dyssynchrony Index
sensitivity of 95% and specificity of 83%, and correlated with reverse remodeling (ⱖ15% reduction of LV ESV) in ischemic and nonischemic patients. The strain delay index constitutes a marker of LV dyssynchrony and residual myocardial contractility and may help overcome the limitations of several currently used time delay indexes that do not take account of residual myocardial contractility. Myocardial deformation is most commonly referenced in the radial, circumferential, and longitudinal coordinate systems. Because LV dyssynchrony is in fact a 3-dimensional phenomenon, it is thought that these 3 types of deformation do not provide the same information in the failing heart. Because speckle-tracking echocardiography uses routine grayscale images, this technique, unlike tissue Doppler imaging, allows the assessment of 3 different types of stain (radial, circumferential, and longitudinal). Therefore, the utility of radial dyssynchrony by speckle-tracking strain for the quantification of LV dyssynchrony and prediction of response to CRT is well recognized.9 –12,16,17 Our study demonstrated that the combined approach of radial, circumferential, and longitudinal SDIs results in a more complete assessment of SDI than the use of individual SDI parameters or of radial dyssynchrony by speckle-tracking strain alone. In particular, all 3 SDIs positively improved the ability to identify patients with a good chance of responding to CRT. Therefore, a comprehensive approach taking into consideration myocardial deformation including radial, circumferential, and longitudinal function may prove to be even more useful for enhancement of the ability to identify patients with a good chance of responding to CRT. The main difference between our SDI and Lim et al’s13 strain delay index is that the strain delay index requires specific software for analysis. Our index, the SDI, can be calculated as the average of the wasted energy per segment due to dyssynchrony, which can be determined by the difference between -peak and -AVC, without the need for specific software, and can be applied to other manufacturers’ speckle-tracking systems. Another difference between these 2 methods is the definition of end-systole. They used the timing of the peak global strain to determine the timing of end-systole, but the timing of end-systole could be difference in a patient by radial, circumferential, and longitudinal strain curve. We determined the timing of the AVC using pulsed Doppler wave of LV outflow as end-systole, and it was recognized as “physiologic end-systole,” including whole cardiac function. In addition, our SDI expressed average of wasted energy, which does not depend on the number of segments. Finally, they assumed to be 0 for a segment that exhibited biphasic longitudinal strain with a peak positive value greater than a maximal absolute negative value. However, we did not assume to be 0 for these segments, because we recognized that the segment with biphasic strain curve was not only dyskinesia but also severe early or delayed contraction, which was the 1 of the important factor of response to CRT. In this study, the prediction of response to CRT in SDI only after AVC was higher than that in SDI only before AVC, but lower than that in our original SDI including both before and after AVC. These findings indicated that the wasted energy after AVC might be important for CRT than that before AVC, because any contraction after AVC did not
445
contribute to cardiac output. Meanwhile, the total wasted energy both before and after AVC might be the most important for CRT. This study covered a small number of patients at a single center, so future studies of larger patient populations are necessary to determine the utility of the SDI for predicting response to CRT. All echocardiographic examinations must be evaluated by an observer blinded to the patients’ characteristics. We also observed a similar utility of radial SDI in patients with wider QRS duration (⬎150 ms; n ⫽ 39), which predicted response to CRT with sensitivity of 82%, specificity of 82%, and an area under the curve of 0.88. In contrast, the utility of SDI in patients with narrower QRS duration (⬍130 ms) was unknown, because we included only 2 patients with QRS durations of 120 to 130 ms in this study. The precise reason for worse sensitivity and specificity than strain delay index observed using our SDI is unknown but may include differences in the general population of the study. Moreover, the strain delay index may be computed automatically with less reproducibility. Limitations of the speckle-tracking method include endocardial border tracing, which requires care when manually finetuning the regions of interest to capture the early septal motion followed by later posterior or lateral wall motion in patients with heart failure with wide QRS complexes and adjustment of its width for dyssynchrony analysis before generating and measuring regional strain. Furthermore, speckle-tracking echocardiography requires experience to achieve sufficiently reproducible results and requires raining.15,29 Another limitation is that only 25% of patients in this study had ischemic cardiomyopathy. A possible reason might be the difference in cause between patients with chronic severe heart failure in Japan and other countries, especially the United States and Europe. Osada et al30 reported that 80% of patients with chronic severe heart failure among 284 applicants for heart transplantation in Japan were diagnosed with idiopathic dilated cardiomyopathy. Acknowledgment: We are grateful for the support of the entire staff of the echocardiography and electrophysiology laboratories of Kobe University Hospital. 1. Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E, Kocovic DZ, Packer M, Clavell AL, Hayes DL, Ellestad M, Trupp RJ, Underwood J, Pickering F, Truex C, McAtee P, Messenger J. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002;346: 1845–1853. 2. Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco T, Carson P, DiCarlo L, DeMets D, White BG, DeVries DW, Feldman AM. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004;350:2140 –2150. 3. Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L, Tavazzi L. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005;352:1539 – 1549. 4. Linde C, Leclercq C, Rex S, Garrigue S, Lavergne T, Cazeau S, McKenna W, Fitzgerald M, Deharo JC, Alonso C, Walker S, Braunschweig F, Bailleul C, Daubert JC. Long-term benefits of biventricular pacing in congestive heart failure: results from the Multisite Stimulation in Cardiomyopathy (MUSTIC) study. J Am Coll Cardiol 2002; 40:111–118.
446
The American Journal of Cardiology (www.ajconline.org)
5. Epstein AE, DiMarco JP, Ellenbogen KA, Estes NA III, Freedman RA, Gettes LS, Gillinov AM, Gregoratos G, Hammill SC, Hayes DL, Hlatky MA, Newby LK, Page RL, Schoenfeld MH, Silka MJ, Stevenson LW, Sweeney MO, Smith SC Jr, Jacobs AK, Adams CD, Anderson JL, Buller CE, Creager MA, Ettinger SM, Faxon DP, Halperin JL, Hiratzka LF, Hunt SA, Krumholz HM, Kushner FG, Lytle BW, Nishimura RA, Ornato JP, Riegel B, Tarkington LG, Yancy CW. ACC/ AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the ACC/AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and Antiarrhythmia Devices) developed in collaboration with the American Association for Thoracic Surgery and Society of Thoracic Surgeons. J Am Coll Cardiol 2008; 51:e1– e62. 6. Bax JJ, Bleeker GB, Marwick TH, Molhoek SG, Boersma E, Steendijk P, van der Wall EE, Schalij MJ. Left ventricular dyssynchrony predicts response and prognosis after cardiac resynchronization therapy. J Am Coll Cardiol 2004;44:1834 –1840. 7. Gorcsan J III, Kanzaki H, Bazaz R, Dohi K, Schwartzman D. Usefulness of echocardiographic tissue synchronization imaging to predict acute response to cardiac resynchronization therapy. Am J Cardiol 2004;93:1178 –1181. 8. Yu CM, Chau E, Sanderson JE, Fan K, Tang MO, Fung WH, Lin H, Kong SL, Lam YM, Hill MR, Lau CP. Tissue Doppler echocardiographic evidence of reverse remodeling and improved synchronicity by simultaneously delaying regional contraction after biventricular pacing therapy in heart failure. Circulation 2002;105:438 – 445. 9. Suffoletto MS, Dohi K, Cannesson M, Saba S, Gorcsan J III. Novel speckle-tracking radial strain from routine black-and-white echocardiographic images to quantify dyssynchrony and predict response to cardiac resynchronization therapy. Circulation 2006;113:960 –968. 10. Tanaka H, Hara H, Saba S, Gorcsan J III. Prediction of response to cardiac resynchronization therapy by speckle tracking echocardiography using different software approaches. J Am Soc Echocardiogr 2009;22:677– 684. 11. Gorcsan J III, Abraham T, Agler DA, Bax JJ, Derumeaux G, Grimm RA, Martin R, Steinberg JS, Sutton MS, Yu CM. Echocardiography for cardiac resynchronization therapy: recommendations for performance and reporting—a report from the American Society of Echocardiography Dyssynchrony Writing Group endorsed by the Heart Rhythm Society. J Am Soc Echocardiogr 2008;21:191–213. 12. Delgado V, Ypenburg C, van Bommel RJ, Tops LF, Mollema SA, Marsan NA, Bleeker GB, Schalij MJ, Bax JJ. Assessment of left ventricular dyssynchrony by speckle tracking strain imaging comparison between longitudinal, circumferential, and radial strain in cardiac resynchronization therapy. J Am Coll Cardiol 2008;51:1944 –1952. 13. Lim P, Buakhamsri A, Popovic ZB, Greenberg NL, Patel D, Thomas JD, Grimm RA. Longitudinal strain delay index by speckle tracking imaging: a new marker of response to cardiac resynchronization therapy. Circulation 2008;118:1130 –1137. 14. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA, Picard MH, Roman MJ, Seward J, Shanewise JS, Solomon SD, Spencer KT, Sutton MS, Stewart WJ. Recommendations for chamber quantification: a report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005;18:1440 –1463. 15. Chung ES, Leon AR, Tavazzi L, Sun JP, Nihoyannopoulos P, Merlino J, Abraham WT, Ghio S, Leclercq C, Bax JJ, Yu CM, Gorcsan J III,
16.
17.
18.
19. 20.
21.
22.
23.
24.
25.
26.
27.
28. 29. 30.
St John Sutton M, De Sutter J, Murillo J. Results of the Predictors of Response to CRT (PROSPECT) trial. Circulation 2008;117: 2608 –2616. Gorcsan J III, Tanabe M, Bleeker GB, Suffoletto MS, Thomas NC, Saba S, Tops LF, Schalij MJ, Bax JJ. Combined longitudinal and radial dyssynchrony predicts ventricular response after resynchronization therapy. J Am Coll Cardiol 2007;50:1476 –1483. Tops LF, Suffoletto MS, Bleeker GB, Boersma E, van der Wall EE, Gorcsan J III, Schalij MJ, Bax JJ. Speckle-tracking radial strain reveals left ventricular dyssynchrony in patients with permanent right ventricular pacing. J Am Coll Cardiol 2007;50:1180 –1188. Ishikawa T, Sumita S, Kimura K, Kikuchi M, Kosuge M, Kuji N, Endo T, Sugano T, Sigemasa T, Kobayashi I, Tochikubo O, Usui T. Prediction of optimal atrioventricular delay in patients with implanted DDD pacemakers. Pacing Clin Electrophysiol 1999;22:1365–1371. van Gelder BM, Bracke FA, Meijer A, Lakerveld LJ, Pijls NH. Effect of optimizing the VV interval on left ventricular contractility in cardiac resynchronization therapy. Am J Cardiol 2004;93:1500 –1503. Perego GB, Chianca R, Facchini M, Frattola A, Balla E, Zucchi S, Cavaglia S, Vicini I, Negretto M, Osculati G. Simultaneous vs. sequential biventricular pacing in dilated cardiomyopathy: an acute hemodynamic study. Eur J Heart Fail 2003;5:305–313. Becker M, Kramann R, Franke A, Breithardt OA, Heussen N, Knackstedt C, Stellbrink C, Schauerte P, Kelm M, Hoffmann R. Impact of left ventricular lead position in cardiac resynchronization therapy on left ventricular remodelling. A circumferential strain analysis based on 2D echocardiography. Eur Heart J 2007;28:1211–1220. Song ZZ. Impact of left ventricular lead position in cardiac resynchronization therapy on left ventricular remodelling: a circumferential strain analysis based on 2D echocardiography. Eur Heart J 2008;29: 684. Ansalone G, Giannantoni P, Ricci R, Trambaiolo P, Fedele F, Santini M. Doppler myocardial imaging to evaluate the effectiveness of pacing sites in patients receiving biventricular pacing. J Am Coll Cardiol 2002;39:489 – 499. Helm RH, Byrne M, Helm PA, Daya SK, Osman NF, Tunin R, Halperin HR, Berger RD, Kass DA, Lardo AC. Three-dimensional mapping of optimal left ventricular pacing site for cardiac resynchronization. Circulation 2007;115:953–961. Bleeker GB, Kaandorp TA, Lamb HJ, Boersma E, Steendijk P, de Roos A, van der Wall EE, Schalij MJ, Bax JJ. Effect of posterolateral scar tissue on clinical and echocardiographic improvement after cardiac resynchronization therapy. Circulation 2006;113:969 –976. Norisada K, Kawai H, Tanaka H, Tatsumi K, Onishi T, Fukuzawa K, Yoshida A, Hirata KI. Myocardial Contractile Function in the Region of the Left Ventricular Pacing Lead Predicts the Response to Cardiac Resynchronization Therapy Assessed by Two-Dimensional Speckle Tracking Echocardiography. J Am Soc Echocardiogr 2010;23:181– 189. Ypenburg C, Schalij MJ, Bleeker GB, Steendijk P, Boersma E, DibbetsSchneider P, Stokkel MP, van der Wall EE, Bax JJ. Impact of viability and scar tissue on response to cardiac resynchronization therapy in ischaemic heart failure patients. Eur Heart J 2007;28:33– 41. Adelstein EC, Saba S. Scar burden by myocardial perfusion imaging predicts echocardiographic response to cardiac resynchronization therapy in ischemic cardiomyopathy. Am Heart J 2007;153:105–112. Yu CM, Bax JJ, Gorcsan J III. Critical appraisal of methods to assess mechanical dyssynchrony. Curr Opin Cardiol 2009;24:18 –28. Osada K, Imaizumi T. Special report from the heart transplant candidate registry committee in Japan. J Heart Lung Transplant 2005;24: 810 – 814.