Skip Navigation

European Journal of Heart Failure 2008 10(3):281-290; doi:10.1016/j.ejheart.2008.02.003
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (16)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by De Boeck, B. W.L.
Right arrow Articles by Cramer, M. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by De Boeck, B. W.L.
Right arrow Articles by Cramer, M. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2008 European Society of Cardiology

Practical and conceptual limitations of tissue Doppler imaging to predict reverse remodelling in cardiac resynchronisation therapy

Bart W.L. De Boecka,*, Mathias Meinea, Geert E. Leendersa, Arco J. Teskea, Harry van Wessela, J. Hans Kirkelsa, Frits W. Prinzenb, Pieter A. Doevendansa and Maarten J. Cramera

a Department of Cardiology, University Medical Centre Utrecht Heidelberglaan 100 3584 CX Utrecht, The Netherlands
b Department of Physiology, University of Maastricht Maastricht, The Netherlands

* Corresponding author. Department of Cardiology, E03.406, University Medical Centre Utrecht, P.O. Box 855500, 3508 GA Utrecht, The Netherlands. Tel.: +31 30 2506176; fax: +31 30 2516396. E-mail address: b.w.l.deboeck{at}umcutrecht.nl (B.W.L. De Boeck).


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 5. Conclusion
 References
 
Background: Recent, conflicting results about the use of tissue Doppler imaging derived (TDI-) asynchrony indices to predict reverse remodelling after cardiac resynchronisation therapy (CRT) have raised questions about their physiological meaning and methodological limitations.

Methods: In 41 patients, baseline TDI-derived septal to lateral delays of peak velocities (TDI-SL), standard deviation of peak velocities over 12 segments (Ts-SD), and peak 2D longitudinal strain (strain-SL) were compared with volumetric response (reduction in end-systolic volume of ≥ 15%) after at least 6 months of CRT. Timing of peak TDI velocities was compared to timing of 2DS velocities and strain-SL. Influence of sample position, transverse motion, and interobserver inconsistency of the chosen peak velocities was assessed. Diagnostic accuracy of TDI-based delays was compared to accuracy of visual and 2D strain-based assessment.

Results: After 7.0±3.2 months of CRT, 24 patients were classified as responders. TDI-SL and Ts-SD were similar between responders and nonresponders at baseline, did not predict response, and were unaffected by CRT. Visual asynchrony scoring and strain-SL were better predictors of response than TDI-SL and Ts-SD. TDI measurements were highly susceptible to sample location and transverse motion components and poorly correlated with the timing of longitudinal contraction. There was a considerably poor agreement between observers with regard to scoring of TDI-SL and Ts-SD.

Conclusion: TDI-based measurements of asynchrony do not appear robust predictors of volume response to CRT.

Key Words: Heart failure • Resynchronisation • Ventricular remodelling • Tissue Doppler imaging • Strain imaging

Received September 18, 2007; Revised January 2, 2008; Accepted February 4, 2008


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 5. Conclusion
 References
 
Cardiac resynchronisation therapy (CRT) has evolved to become an established therapy for a subset of patients with drug refractory heart failure. CRT improves symptoms, ventricular function, and prognosis by restoration of a more coordinated contraction pattern. However, multiple large trials have indicated a clinical response in only about 70% of patients, including a noticeable placebo effect [1,2]. Reverse remodelling, in particular a reduction of left ventricular end-systolic volume (LVESV), heralds an improved prognosis following CRT but is observed in only 50-65% of patients [3-6]. Failure to respond to resynchronisation has traditionally been attributed to the absence of mechanical asynchrony despite electrical activation delay on the standard electrocardiogram. Accordingly, attention has increasingly shifted towards echocardiographic techniques quantifying mechanical asynchrony. Previous retrospective studies with tissue Doppler imaging (TDI) have suggested a good agreement between TDI-derived asynchrony parameters, more specifically delays between peak velocities of longitudinal motion, and acute and long-term echocardiographic response in patients eligible for CRT [5,7-10]. These encouraging results have prompted many cardiologists to consider TDI-based asynchrony as a potential selection criterion, instigating a movement towards their incorporation in the guidelines [11]. However, the idea of echocardiography-guided patient selection itself is still under consideration and some negative reports have appeared recently [12-14]. A poor predictive performance may have several causes: the ability of the biomarker to depict the underlying pathophysiological substrate may be poor and/or the measurement may be user dependent. With regard to the first issue, we hypothesized that in the asynchronous ventricle timing of peak velocities along the ultrasound beam would unsatisfactorily reflect timing of local myocardial shortening [15]. With regard to the second issue, we hypothesized that variations in the sample position within the basal segment and improper alignment of the ultrasound beam with the ventricular wall could have more than subtle effects on the TDI signal characteristics. In addition, noticeable differences in the measurement result between different observers could also result from inconsistencies in the choice of peak velocities. The present study was designed to prospectively evaluate the aforementioned issues and the overall predictive value of TDI with regard to reverse remodelling.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 5. Conclusion
 References
 
2.1. Study population
Patients with heart failure eligible for CRT (NYHA classification ≥3 despite optimal medical treatment, ejection fraction (LVEF) <35%, and QRS duration ≥130 ms with left bundle branch morphology) were consecutively enrolled in the study. Patients with decompensated heart failure, severe aortic stenosis, intractable ventricular arrhythmias, atrial fibrillation with insufficient rate control, recent (<6 months) myocardial infarction or coronary intervention, were excluded. Predefined criteria to exclude patients from the data analysis consisted of insufficient image quality for reliable calculation of left ventricular volumes and any arrhythmia or device related problem resulting in biventricular pacing for ≤85% of the time. Clinical status (NYHA class), brain-type natriuretic peptide (BNP), and echocardiographic characteristics were prospectively assessed before CRT and after a follow-up of at least 6 months. The study was approved by the local Medical Ethics Committee and conformed with the principles outlined in the Declaration of Helsinki on research in human subjects.

2.2. Echocardiographic protocol
Pulsed and colour Doppler data, colour tissue Doppler, and 2D echocardiographic data were acquired on a Vivid 7 ultrasound machine (General Electric, Milwaukee, USA) using a 3.5 MHz phased array probe. Doppler of the aortic flow was used for timing issues, with systole being defined as the time period between aortic valve opening and closure. The interventricular mechanical delay (IVMD), the left (LVPEP) and right ventricular (RVPEP) pre-ejection period were measured as previously described [11]. Colour-coded TDI was performed from the apical views at frame rates >100 fps. A minimum of 3 loops were acquired at end-expiration and analysed off-line (Echopac version 6.0.1, General Electric, USA). In the apical 4-chamber TDI images, a fixed region of interest (ROI) of 6x6 mm was placed in the basal part of the septal and lateral wall 1-3 mm above the mitral annulus in end-systole to calculate the septal to lateral delay (TDI-SL) in peak systolic velocity (Fig. 1) [5,8]. Time to peak velocity (T1) was measured from the onset of the QRS-complex to the (highest) peak systolic velocity, omitting velocities within the isovolumic periods [7-11]. Time to peak systolic velocities in 6 basal and 6 mid segments was automatically detected by tissue synchronization imaging (TSI) from which the standard deviation over 12 segments (Ts-SD) was determined [10,16]. The cut-offs of TDI-SL>60 ms and Ts-SD>32 ms, as proposed in the PROSPECT study design, were adopted to prospectively predict response to CRT [8,11,17]. Response at follow-up was determined in terms of LV reverse remodelling, patients with a decrease in LVESV of at least 15% were classified as responders. Left ventricular end-systolic (LVESV) and end-diastolic (LVEDV) volume and LVEF were measured by the biplane Simpson's method. To avoid errors due to between-plane asynchrony, special care was taken to perform endocardial border tracing in both planes in exactly matched phases of the isovolumic periods. Mitral insufficiency was quantified (effective regurgitant orifice MRero) [18] and the rate of LV isovolumic pressure augmentation dP/dt deducted from the continuous Doppler mitral regurgitation signal. To determine intraobserver variability, measurements of IVMD, TDI-SL, Ts-SD, strain-SL, and biplane volumes were repeated on a random sample of 40 recordings at least 6 weeks after the initial evaluation.


Figure 01
View larger version (46K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Sample position within the basal segment can affect the measured TDI-SL. Left panel shows the basal and midbasal position of ROI's as defined in the study, with the corresponding traces (see text). A biphasic systolic signal with a late peak is seen in the basal lateral wall (top right, cyan curve), resulting in a TDI-SL of 117 ms. At the midbasal level, the velocity pattern at the lateral wall is less biphasic and displays an early peak (bottom right, green curve); a TDI-SL of –22 ms is measured. The patient responded to CRT.

 
The longitudinal deformation of the septal and the lateral wall was calculated off-line from high resolution single wall B-mode images using 2D strain echocardiography (2DS) based on speckle tracking software (Echopac 2DS version 59) [19]. To represent the integral of longitudinal shortening calculated over the entire wall length, the ROI was set along the endocardial border from base to apex. The timing of the first peak negative value [20] of the longitudinal deformation along the entire ROI was assessed for calculation of the strain-based septal to lateral delay (strain-SL).

2.3. Additional echocardiographic measurements
2.3.1. Variation in position of the sample area
Extra TDI-ROIs were placed in the midbasal segments of all 41 colour-coded 4-chamber images for comparison of basal and midbasal T1 values and SL-delays (Fig. 1). Midbasal was defined as the middle of the basal segment in end-systole.

2.3.2. Influence of transverse motion on TDI measurements
Whereas 2DS can measure true longitudinal velocities along the ventricular wall, TDI-derived signals are composed of the vector sum of longitudinal and radial velocities projected on the ultrasound beam direction. Therefore TDI is susceptible to misalignment, but only if additional radial motion characteristics differ substantially from the longitudinal [21]. Evidence for a combined effect of both factors confounding the "longitudinal" velocities assessed by TDI was checked by comparing the correlation between TDI- and 2DS-derived peak velocity patterns obtained at 28 matched positions in perfectly aligned segments (i.e. reference correlation) with the correlation between these two techniques obtained at matched positions in 28 basal segments.

2.3.3. Consistency of measurement outcome between experts
To test the interobserver agreement in the perceived definition of "peak systolic velocity" we distributed handouts of velocity traces of the first 18 consecutive patients to nine faculty members of an international echocardiography congress dedicated to TDI and CRT. Handouts included still frames displaying the ROI markers in the basal septal and lateral segment, the associated velocity traces of two consecutive beats, and timing event markers. Observers were asked to mark the septal and lateral velocity peaks they would chose in their everyday clinical practice. TDI-SL for all submitted traces was calculated to determine interobserver variability.

2.3.4. Visual impression of asynchrony
Two experienced observers visually scored asynchrony based on a qualitative estimation of the presence and severity of a rocking motion of the apex and/or a septal "whiplash motion" (brisk early systolic inward motion followed by dyskinaesia) [22]. The observers were only allowed to view the available 2D ultrasound data and recorded the presence of relevant asynchrony on a yes or no scale.

2.4. CRT device implantation
Devices were implanted using a single left pectoral incision with transvenous insertion of the LV pacing lead into a side branch of the coronary sinus in 40 patients and on the ventricular free wall by video assisted thoracoscopy in one. The LV-lead was placed midlateral or midposterolateral in 34 patients (82.9%), midposterior in 4 (9.8%), apicolateral in 2 (4.9%), and basolateral in 1 (2.4%). Right atrial and right ventricular leads were positioned in the atrial appendage and apex respectively. In 31 patients (76%) the AV and VV delay were acutely optimized by invasive measurements aiming at the highest maximal rate of LV pressure change (dP/dtmax) (Radi, Uppsala, Sweden). Devices were optimized based on the intracardiac electrogram (EGM) in the remaining 10 patients (24%) [23].

2.5. Statistical analysis
Statistical analysis was performed using SPSS version 11.5 (SPSS Inc., Chicago, Illinois). Continuous variables were described as mean±SD, frequencies and percentages were calculated for categorical variables. A p-value <0.05 was considered statistically significant. Intraobserver variability was assessed by the coefficient of variation. Interobserver reproducibility of TDI-SL measurements between the 9 experts was evaluated by the intra-class correlation coefficient (ICC). Paired measurements (basal versus midbasal TDI, TDI- versus 2DS-derived velocities) were analysed according to Bland-Altman to assess limits of agreement and the difference (bias) was controlled by t-test. Differences in continuous and categorical variables between responders and nonresponders were assessed by independent Students t-test and Fischer's exact test, respectively. Differences between echocardiographic characteristics at baseline and follow-up were evaluated using the paired Students t-test. Correlation between TDI-SL and strain-SL asynchrony and between baseline asynchrony parameters and changes in LVESV at follow-up was expressed by Pearson's or Spearman's correlation coefficients where appropriate. Receiver operating characteristics (ROC) curves were constructed and the area under the curve (AUC) was measured to evaluate the predictive value of the baseline variables. Sensitivity, specificity, positive (PPV) and negative predictive value (NPV), and accuracy of response prediction by the predefined cut-off points for TDI-SL, Ts-SD, and for the visual and strain assessment of asynchrony were calculated. Finally, a forward stepwise logistic regression analysis was performed to test the independent predictive value of baseline patient and echocardiographic characteristics with regard to volumetric response. The model included age, sex, type of the underlying disease, rhythm, QRS duration, NYHA class, BNP, LVEF, LVESV, MRero, IVMD, TDI-SL, Ts-SD, and strain-SL (LVPEP was removed as it is collinear with IVMD). Results are expressed in odds ratio (OR), 95% confidence intervals (CI), and p-value.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 5. Conclusion
 References
 
3.1. Study population
Of the 46 patients screened, four were excluded due to poor image quality and one because of sustained slow ventricular tachycardia after CRT. Forty-one patients were therefore included in the final population. Baseline characteristics of the total population and differences between baseline characteristics of responders and nonresponders are presented in Table 1. Aetiology of heart failure was ischaemic in 39% of the patients; there were no differences in baseline characteristics between ischaemic and non-ischaemic patients. Pacemaker implantation was successful in all patients; no procedure-related complications occurred during follow-up. Follow-up was completed for all patients after 7.0±3.2 months of CRT. One patient underwent heart transplantation and one died 6.5 and 10 months after CRT respectively. Both patients had been classified as volumetric nonresponders after 6 months.


View this table:
[in this window]
[in a new window]

  		Table 1 Baseline demographic and clinical patient characteristics

 
3.2. Response to CRT
After a mean of 7.0±3.2 months of follow-up, CRT induced reverse remodelling (LVEDV 238±82 ml to 199±93 ml, p<0.001; LVESV 197±78 ml to 149±90 ml, p<0.001), improved LV function (LVEF 18.6±6.9% to 29.1±12.7%, p<0.001; dP/dt 545.9±171 mmHg/s to 745.9±221 mmHg/s, p<0.001), and decreased the severity of mitral regurgitation (MRero 10.2±8.5 mm2 to 6.4±7.1 mm2, p=0.001) in the total population. In addition, resynchronisation improved heart failure symptoms (NYHA 3.2±0.3 to 2.2±0.8, p<0.001) and tended to reduce the plasma levels of BNP (182±178 pmol/ml to 139±219 pmol/ml, p=0.095). Interventricular asynchrony decreased (IVMD 56±25 ms to 6±27 ms, p<0.001) and strain analysis indicated intraventricular resynchronisation by CRT (strain-SL 262±130 ms to 3±137 ms, p<0.001). TDI-estimates of intraventricular asynchrony did not improve (TDI-SL 30.1±57.3 ms to 18.1±62.5 ms, p=0.358 and Ts-SD 44.1±15.6 ms to 41.1±16.5 ms, p=0.308). None of these effects of CRT differed between ischaemic and non-ischaemic patients, or between patients optimized invasively or by the EGM.

Of the total population, 24 patients (58.5%) displayed a reduction in LVESV ≥15% and were classified as responders. Clinical and echocardiographic differences between responders and nonresponders at baseline and at follow-up are presented in Tables 1 and 2. At baseline, nonresponders on average had more severe clinical and echocardiographic evidence of LV dysfunction and remodelling, a strong trend (p=0.054) towards more mitral insufficiency and significantly less inter- and intraventricular asynchrony assessed by IVMD, LVPEP and strain-SL. In contrast, Ts-SD and TDI-SL did not discriminate responders from nonresponders at baseline and failed to indicate any resynchronisation at follow-up in either group. The amount of residual interventricular and strain-based intraventricular asynchrony at follow-up was also limited in nonresponders and did not differ from the residual asynchrony in responders.


View this table:
[in this window]
[in a new window]

  		Table 2 Echocardiographic differences between responders and nonresponders at baseline and at follow-up

 
Volumetric responders displayed both a marked clinical response (NYHA 3.1±0.2 to 1.8±0.5, p<0.001) as well as a prominent decrease in BNP (141±134 pmol/ml to 45±46 pmol/ml, p=0.002). In contrast, BNP levels failed to decrease at 6 months of follow-up in patients classified as nonresponders (237±215 pmol/ml to 264±288 pmol/ml) despite minor improvements in echocardiographic parameters and in functional status (NYHA 3.2±0.4 to 2.8±0.9, p=0.041, 10/17 patients reporting improvement of at least 1 NYHA class).

3.3. Predictive value and correlation of asynchrony parameters with reverse remodelling
Baseline IVMD, LVPEP and strain-SL, but none of the TDI-based indices, significantly correlated with the magnitude of the volumetric response and revealed discriminatory power between responders and nonresponders at ROC analysis (Table 3). The sensitivity, specificity and predictive value of the previously proposed cut-off values of TDI-SL and Ts-SD [11] was poor and was outperformed by global markers of asynchrony, by the visual asynchrony assessment of both observers, and by a judgment based on strain-delays (Table 4). At multivariate analysis baseline asynchrony assessed by strain-SL (OR 1.019; 95% CI 1.005-1.033; p=0.008) and IVMD (OR 1.053; 95% CI 1.000-1.108; p=0.049) as well as baseline remodelling assessed by LVESV (OR 0.974; 95% CI 0.953-0.995; p=0.016) were identified as independent predictors of volumetric response.


View this table:
[in this window]
[in a new window]

  		Table 3 Baseline asynchrony predictors of reverse remodelling

 


View this table:
[in this window]
[in a new window]

  		Table 4 Sensitivity, specificity and predictive values of echocardiographic cut-off values and of the visual assessment to predict volumetric response to CRT

 
3.4. Intraobserver variability of the main asynchrony and end point parameters
Acquiring adequate strain data at the lateral wall was challenging in 4 patients but none of the data was rejected from the analysis. The coefficients of intraobserver variation (CV) were 11.3% for TDI-SL, 12.9% for Ts-SD, 5.6% for LVPEP, 13.5% for IVMD and 8.9% for strain-SL in this study. The intraobserver reliability of the biplane LVESV and LVEF were reasonable at a CV of 5.6% and 5.8% respectively.

3.5. Technical and interobserver factors influencing the accuracy and results of longitudinal velocity measurements
3.5.1. Variation in position of the sample area
The measured TDI-SL was significantly higher (p<0.01) when the region of interest was placed in the basal part of the basal segment compared to the middle of part this segment. The wide limits of agreement indicate a major impact on the measurement by varying the sample position by only 1.5-2 cm within this segment (Figs. 1 and 2).


Figure 02
View larger version (17K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Bland-Altman plot with 95% agreement boundaries between TDI-SL acquired in the lower (basal) and middle part (midbasal) of the basal segments.

 
3.5.2. Influence of transverse motion
Comparison between TDI-derived and 2DS-derived timing measurements showed a good correlation in segments perfectly aligned to the ultrasound beam (r=0.927, p<0.001) whereas in basal segments the measurements correlated significantly less (r=0.464, p=0.013) than the former reference correlation (p<0.01 for comparison of the correlation). Moreover, Bland-Altman revealed a nearly 3-fold higher variability between the 2 techniques in basal segments (95% LOA 165 ms) compared to the reference in aligned segments (95% LOA 58 ms). These findings strongly suggest that the TDI-velocity pattern in the former less optimally aligned segments included radial velocity components on top of the true longitudinal velocity pattern.

3.5.3. Time to peak longitudinal velocities versus time to peak shortening
Neither for the basal nor for the midbasal segments any correlation was found between timing of septal to lateral delay of peak contraction by 2DS and the corresponding delay of peak velocities obtained by TDI (all r<0.3, all p>0.2).

3.5.4. Interobserver inconsistencies in the definition of peak velocities
Of the 18 cases evaluated by the nine experts, there was full agreement on the TDI-SL in only three patients. In seven cases, there was an agreement between 8 out of the 9 observers, with no single observer who systematically scored differently from the others. Two experts did not measure any TDI-SL in 2 traces, due to the absence of any positive systolic peak in the septum. The intra-class correlation coefficient was 0.42, indicating a moderate inter-rater reliability.


   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
5. Conclusion
 References
 
In the present study, two commonly used and advocated TDI-based methods of intraventricular asynchrony assessment [11] failed to prospectively predict reverse remodelling following CRT and had no apparent added value over global echocardiographic parameters of asynchrony (LVPEP and IVMD) or the experienced eye. The poor performance of TDI can at least partly be explained by a combination of inherent shortcomings of the TDI technique and poorly defined conventions about acquisition and measurement methodology.

4.1. Comparison with previous TDI studies
The overall response and responder rate was as can be expected in this population, even though the proportion of patients with NYHA IV symptoms was fairly large and the LVEF rather low compared to most reported TDI studies. Nevertheless, we were unable to reproduce the predictive value of colour-coded TDI reported in the recent literature (for overview: see Ref. [12]). A few negative reports have recently surfaced for pulsed TDI [12-14]. Among them, the most compelling evidence has come from a prospective multi-centre Italian study of 133 patients in which pulsed TDI-SL failed to predict clinical as well as echocardiographic response [12]. Because multiple segment models of asynchrony have been suggested to yield superior results both for pulsed and colour-coded TDI compared to a septal and lateral assessment only, we also evaluated Ts-SD [9,10]. However, in our population this approach poorly predicted volumetric remodelling after CRT [14]. According to a recent congress presentation, the multi-centre PROSPECT study also raised questions about the feasibility and accuracy of TDI. The publication of the results from this study is pending [24].

Simple global parameters of inter- and intraventricular asynchrony, as well as visual signs of contractile asynchrony, were of value in our study as well as in previous larger studies [12,22,25,26] and should therefore probably not be abandoned in favour of local TDI measurements but rather analysed in addition to them.

4.2. Potential explanations for the present findings
4.2.1. Limited value of TDI to represent contractile asynchrony in dilated hearts
Our study indicates a considerable degree of variability in both the TDI signal characteristics as well as their interpretation by experts. We often encountered complex and multiphasic systolic TDI signals rendering a uniform interpretation more difficult. Other important confounding factors appear to be the susceptibility of the signals to sample position and to the alignment of the ultrasound beam with the ventricular wall. The latter can be understood if one considers that myocardial deformation is a complex three-dimensional process with different amounts and timing of shortening in different directions [21]. Consequently, transverse motion components in Doppler signals of poorly aligned wall segments may result in erroneous peak velocities.

More importantly, asynchrony indices in this study, which were based on known physiological principles of contractile asynchrony, predicted response to CRT even when assessed visually. Despite the presence of left bundle branch block in all patients, TDI-SL indicated a delayed contraction of the lateral wall in only 66% (versus 95% by strain-SL) [15]. The discrepancy between these two indices may be particularly large in asynchronous ventricles because the very nature of asynchronous contraction renders the myocardial forces unbalanced throughout the ventricle and around the LV apex [27]. The result of this interaction is variable but is often visible as a rocking or rotational motion of the apex and/or a "whiplash motion" of the septum [22]. Under such circumstances, the motion pattern at the base of the ventricle not only represents the integral of longitudinal shortening between apex and base but also the highly variable effect of apical and global ventricular motion (Fig. 3). A large part of the poor predictive value of TDI can therefore be attributed to its aforementioned practical and conceptual limitations for depicting contractile asynchrony.


Figure 03
View larger version (86K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Measurement method and relation between TDI-SL and strain-SL. In this CRT responder, peak systolic velocities assessed by TDI occur simultaneously (red arrows upper panel, TDI-SL: 5 ms). Strain of the entire wall is measured by 2DS (lower panels). Septal contraction (lower left quadrant) starts early and has an early peak, while the lateral wall (lower right quadrant) displays prestretch, a late onset and peak. Strain-SL was 315 ms.

 
4.2.2. Evidence for absence of reverse remodelling despite resynchronisation? A "beyond repair"-hypothesis
Although the multifaceted nature of asynchrony makes it unlikely for any single asynchrony parameter to reliably predict response to CRT, another contributing factor may be that some hearts are "beyond repair", even though they are asynchronous. In accordance with previous prospective, multi-centre CRT trials and other interventional studies in advanced heart failure, the nonresponders in our study were characterised as a group by more severe heart failure and more remodelled ventricles [3,12,28,29]. These findings support the idea that some ventricles are beyond repair regardless of the presence of an amendable substrate. In fact in nine of our nonresponders, marked baseline asynchrony was present (strain-SL >150 ms) (Supplementary figure — see Appendix). In all but one of them, invasive dP/dtmax during the procedure increased by ≥25% [30,31]. Moreover, at follow-up neither lead location nor V-V intervals or pacing percentage (96%) differed between responders and nonresponders. Inadequately delivered CRT therefore seems a less plausible cause for nonresponse in most of these patients. More studies are warranted on this topic.

4.3. Limitations
This study has the limitations inherent to single centre, observational studies with relatively small numbers of patients. Because of the small numbers, results from the subgroup analysis in particular should be interpreted with caution. Nonetheless, applying strict entry criteria to minimize the confounding effects of poor image quality or inadequate therapy, we reproduced the majority of echocardiographic findings from recently conducted, larger, multi-centre prospective trials [12,28].

In advanced heart failure, improving both symptoms and prognosis are important treatment goals. However, in the present echocardiographic study, reverse remodelling was chosen as the primary outcome parameter. Volume changes can be assessed more objectively than symptomatic improvements (NYHA class) and have a stronger impact on the prognosis [2,6]. Monitoring of natriuretic peptides has recently been advocated as an alternative objective method to assess clinical improvement and prognosis [2,32]. However, a universal definition of CRT responders is still lacking. In the present study, we classified patients into nonresponders based on the failure to achieve marked reverse remodelling (≥15% reduction in LVESV) ±6 months after CRT. This was mainly done to allow comparison with previous TDI studies. We realise that such dichotomous classification ignores the potential benefits of less pronounced reverse remodelling or of disease stabilization [33]. In fact, minor clinical and echocardiographic improvements also occurred in our "nonresponders". We therefore tried to circumvent the problem of dichotomization by performing correlation analysis for the main parameters. In this respect, an important finding was that the strongest predictors of volumetric response in the present study also correlated with decreases in (log transformed) BNP (correlation for strain-SL r=0.457, p=0.004, respectively for IVMD r=0.547, p<0.001).

Finally, the effects of CRT on ventricular structure and function have been demonstrated to be dynamic over time [2,33]. Therefore our 6-month measurements do not necessarily represent the true long-term benefit of this therapy. Nevertheless, both marked left ventricular reversed remodelling as well as a decrease in natriuretic peptides early after CRT can be regarded as important prognostic factors [2,6,32].

Only IVMD, LVPEP and colour tissue Doppler-derived TDI-SL and Ts-SD were investigated in our study, although many other indices have been proposed and are prospectively evaluated in the PROSPECT-trial [11,17]. This trial will undoubtedly enable more definite conclusions about the value of these parameters [17,24].

We chose the integral of longitudinal deformation assessed by 2DS over the entire length of the septum and the lateral wall for a better comparison with velocities at the base. The length over which the deformation was calculated and the high image resolution of single wall recordings, enabled measurements of strain-SL in all patients and with reasonable intraobserver variability. In addition, this approach approximates the impact of asynchrony by including a relevant proportion of the ventricle in the analysis [19]. Asynchrony of circumferential contraction would theoretically yield a higher predictive value [21], but in our experience is technically less feasible.


   5. Conclusion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 5. Conclusion
References
 
In the present study, TDI-based asynchrony measures appeared to be poor predictors of reverse remodelling after CRT. As potential explanations, we identified the critical dependency of TDI on properly agreed standards of acquisition and measurement methodology and its inherent limited ability to represent the contractile sequence in dilated ventricles.


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 5. Conclusion
 References
 

  1. Pires L.A., Abraham W.T., Young J.B., Johnson K.M. Clinical predictors and timing of New York Heart Association class improvement with cardiac resynchronization therapy in patients with advanced chronic heart failure: results from the Multicenter InSync Randomized Clinical Evaluation (MIRACLE) and Multicenter InSync ICD Randomized Clinical Evaluation (MIRACLE-ICD) trials. Am Heart J (2006) 151:837–843.[CrossRef][Web of Science][Medline]
  2. Fruhwald F.M., Fahrleitner-Pammer A., Berger R., et al. Early and sustained effects of cardiac resynchronization therapy on N-terminal pro-B-type natriuretic peptide in patients with moderate to severe heart failure and cardiac dyssynchrony. Eur Heart J (2007) 28:1592–1597.[Abstract/Free Full Text]
  3. Stellbrink C., Breithardt O.A., Franke A., et al. Impact of cardiac resynchronization therapy using hemodynamically optimized pacing on left ventricular remodeling in patients with congestive heart failure and ventricular conduction disturbances. J Am Coll Cardiol (2001) 38:1957–1965.[Abstract/Free Full Text]
  4. Pitzalis M.V., Iacoviello M., Romito R., et al. Cardiac resynchronization therapy tailored by echocardiographic evaluation of ventricular asynchrony. J Am Coll Cardiol (2002) 40:1615–1622.[Abstract/Free Full Text]
  5. Bax J.J., Bleeker G.B., Marwick T.H., et al. Left ventricular dyssynchrony predicts response and prognosis after cardiac resynchronization therapy. J Am Coll Cardiol (2004) 44:1834–1840.[Abstract/Free Full Text]
  6. Yu C.M., Bleeker G.B., Fung J.W., et al. Left ventricular reverse remodeling but not clinical improvement predicts long-term survival after cardiac resynchronization therapy. Circulation (2005) 112:1580–1586.[Abstract/Free Full Text]
  7. Yu C.M., Chau E., Sanderson J.E., et al. 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.[Abstract/Free Full Text]
  8. Bax J.J., Molhoek S.G., van Erven L., et al. Usefulness of myocardial tissue Doppler echocardiography to evaluate left ventricular dyssynchrony before and after biventricular pacing in patients with idiopathic dilated cardiomyopathy. Am J Cardiol (2003) 91:94–97.[CrossRef][Web of Science][Medline]
  9. Yu C.M., Fung J.W., Zhang Q., et al. Tissue Doppler imaging is superior to strain rate imaging and postsystolic shortening on the prediction of reverse remodeling in both ischemic and nonischemic heart failure after cardiac resynchronization therapy. Circulation (2004) 110:66–73.[Abstract/Free Full Text]
  10. Yu C.M., Zhang Q., Fung J.W., et al. A novel tool to assess systolic asynchrony and identify responders of cardiac resynchronization therapy by tissue synchronization imaging. J Am Coll Cardiol (2005) 45:677–684.[Abstract/Free Full Text]
  11. Bax J.J., Ansalone G., Breithardt O.A., et al. Echocardiographic evaluation of cardiac resynchronization therapy: ready for routine clinical use? A critical appraisal. J Am Coll Cardiol (2004) 44:1–9.[Abstract/Free Full Text]
  12. Achilli A., Peraldo C., Sassara M., et al. Prediction of response to cardiac resynchronization therapy: the selection of candidates for CRT (SCART) study. Pacing Clin Electrophysiol (2006) 29(Suppl_2):S11–S19.[CrossRef][Medline]
  13. Soliman O.I., Theuns D.A., Geleijnse M.L., et al. Spectral pulsed-wave tissue Doppler imaging lateral-to-septal delay fails to predict clinical or echocardiographic outcome after cardiac resynchronization therapy. Europace (2007) 9:113–118.[Abstract/Free Full Text]
  14. Porciani M.C., Lilli A., Macioce R., et al. Utility of a new left ventricular asynchrony index as a predictor of reverse remodelling after cardiac resynchronization therapy. Eur Heart J (2006) 27:1818–1823.[Abstract/Free Full Text]
  15. Breithardt O.A., Stellbrink C., Herbots L., et al. Cardiac resynchronization therapy can reverse abnormal myocardial strain distribution in patients with heart failure and left bundle branch block. J Am Coll Cardiol (2003) 42:486–494.[Abstract/Free Full Text]
  16. Van de Veire N.R., Bleeker G.B., De Sutter J., et al. Tissue synchronisation imaging accurately measures left ventricular dyssynchrony and predicts response to cardiac resynchronisation therapy. Heart (2007) 93:1034–1039.[Abstract/Free Full Text]
  17. Yu C.M., Abraham W.T., Bax J., et al. Predictors of response to cardiac resynchronization therapy (PROSPECT)—study design. Am Heart J (2005) 149:600–605.[CrossRef][Web of Science][Medline]
  18. Dujardin K.S., Enriquez-Sarano M., Bailey K.R., Nishimura R.A., Seward J.B., Tajik A.J. Grading of mitral regurgitation by quantitative Doppler echocardiography: calibration by left ventricular angiography in routine clinical practice. Circulation (1997) 96:3409–3415.[Abstract/Free Full Text]
  19. Teske A.J., De Boeck B.W., Melman P.G., Sieswerda G.T., Doevendans P.A., Cramer M.J. Echocardiographic quantification of myocardial function using tissue deformation imaging — a guide to image acquisition and analysis using tissue Doppler and speckle tracking. Cardiovasc Ultrasound (2007) 5:27.[CrossRef][Medline]
  20. Zwanenburg J.J., Gotte M.J., Marcus J.T., et al. Propagation of onset and peak time of myocardial shortening in time of myocardial shortening in ischemic versus nonischemic cardiomyopathy: assessment by magnetic resonance imaging myocardial tagging. J Am Coll Cardiol (2005) 46:2215–2222.[Abstract/Free Full Text]
  21. Helm R.H., Leclercq C., Faris O.P., et al. Cardiac dyssynchrony analysis using circumferential versus longitudinal strain: implications for assessing cardiac resynchronization. Circulation (2005) 111:2760–2767.[Abstract/Free Full Text]
  22. Jansen A.H., van Dantzig J., Bracke F., et al. Qualitative observation of left ventricular multiphasic septal motion and septal-to-lateral apical shuffle predicts left ventricular reverse remodeling after cardiac resynchronization therapy. Am J Cardiol (2007) 99:966–969.[CrossRef][Web of Science][Medline]
  23. Baker J.H. 2nd, McKenzie J. 3rd, Beau S., et al. Acute evaluation of programmer-guided AV/PV and VV delay optimization comparing an IEGM method and echocardiogram for cardiac resynchronization therapy in heart failure patients and dual-chamber ICD implants. J Cardiovasc Electrophysiol (2007) 18:185–191.[CrossRef][Web of Science][Medline]
  24. Cleland J.G., Abdellah A.T., Khaleva O., Coletta A.P., Clark A.L. Clinical trials update from the European Society of Cardiology Congress 2007: 3CPO, ALOFT, PROSPECT and statins for heart failure. Eur J Heart Fail (2007) 9:1070–1073.[Abstract/Free Full Text]
  25. Duncan A.M., Lim E., Clague J., Gibson D.G., Henein M.Y. Comparison of segmental and global markers of dyssynchrony in predicting clinical response to cardiac resynchronization. Eur Heart J (2006) 27:2426–2432.[Abstract/Free Full Text]
  26. Richardson M., Freemantle N., Calvert M.J., Cleland J.G., Tavazzi L. Predictors and treatment response with cardiac resynchronization therapy in patients with heart failure characterized by dyssynchrony: a pre-defined analysis from the CARE-HF trial. Eur Heart J (2007) 28:1827–1834.[Abstract/Free Full Text]
  27. Prinzen F.W., Hunter W.C., Wyman B.T., McVeigh E.R. Mapping of regional myocardial strain and work during ventricular pacing: experimental study using magnetic resonance imaging tagging. J Am Coll Cardiol (1999) 33:1735–1742.[Abstract/Free Full Text]
  28. Diaz-Infante E., Mont L., Leal J., et al. Predictors of lack of response to resynchronization therapy. Am J Cardiol (2005) 95:1436–1440.[CrossRef][Web of Science][Medline]
  29. Bax J.J., Schinkel A.F., Boersma E., et al. Extensive left ventricular remodeling does not allow viable myocardium to improve in left ventricular ejection fraction after revascularization and is associated with worse long-term prognosis. Circulation (2004) 110:II18–II22.[Web of Science][Medline]
  30. Nelson G.S., Curry C.W., Wyman B.T., et al. Predictors of systolic augmentation from left ventricular preexcitation in patients with dilated cardiomyopathy and intraventricular conduction delay. Circulation (2000) 101:2703–2709.[Abstract/Free Full Text]
  31. Helm R.H., Byrne M., Helm P.A., et al. Three-dimensional mapping of optimal left ventricular pacing site for cardiac resynchronization. Circulation (2007) 115:953–961.[Abstract/Free Full Text]
  32. Olsson L.G., Swedberg K., Cleland J.G., et al. Prognostic importance of plasma NT-pro BNP in chronic heart failure in patients treated with a beta-blocker: results from the Carvedilol Or Metoprolol European Trial (COMET) trial. Eur J Heart Fail (2007) 9:795–801.[Abstract/Free Full Text]
  33. Sutton M.G., Plappert T., Hilpisch K.E., Abraham W.T., Hayes D.L., Chinchoy E. Sustained reverse left ventricular structural remodeling with cardiac resynchronization at one year is a function of etiology: quantitative Doppler echocardiographic evidence from the Multicenter InSync Randomized Clinical Evaluation (MIRACLE). Circulation (2006) 113:266–272.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Eur J Heart FailHome page
B. W.L. De Boeck, A. J. Teske, M. Meine, G. E. Leenders, M. J. Cramer, F. W. Prinzen, and P. A. Doevendans
Septal rebound stretch reflects the functional substrate to cardiac resynchronization therapy and predicts volumetric and neurohormonal response
Eur J Heart Fail, September 1, 2009; 11(9): 863 - 871.
[Abstract] [Full Text] [PDF]

Home page
 J Am Coll CardiolHome page
N. M. Hawkins, M. C. Petrie, M. I. Burgess, and J. J.V. McMurray
Selecting patients for cardiac resynchronization therapy: the fallacy of echocardiographic dyssynchrony.
J. Am. Coll. Cardiol., May 26, 2009; 53(21): 1944 - 1959.
[Abstract] [Full Text] [PDF]

Home page
 J Am Coll Cardiol ImgHome page
O. A. Breithardt
Echocardiographic patient selection for cardiac resynchronization therapy betting on a dead horse?
J. Am. Coll. Cardiol. Img., May 1, 2009; 2(5): 544 - 547.
[Full Text] [PDF]

Home page
 Eur J Heart FailHome page
D. Mele, T. Toselli, F. Capasso, G. Stabile, M. Piacenti, M. Piepoli, S. Giatti, C. Klersy, L. Sallusti, and R. Ferrari
Comparison of myocardial deformation and velocity dyssynchrony for identification of responders to cardiac resynchronization therapy
Eur J Heart Fail, April 1, 2009; 11(4): 391 - 399.
[Abstract] [Full Text] [PDF]

Home page
 Eur J Heart FailHome page
A. D'Andrea, P. Caso, R. Scarafile, L. Riegler, G. Salerno, F. Castaldo, R. Gravino, R. Cocchia, L. Del Viscovo, G. Limongelli, et al.
Effects of global longitudinal strain and total scar burden on response to cardiac resynchronization therapy in patients with ischaemic dilated cardiomyopathy
Eur J Heart Fail, January 1, 2009; 11(1): 58 - 67.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
B. Kirn, A. Jansen, F. Bracke, B. van Gelder, T. Arts, and F. W. Prinzen
Mechanical discoordination rather than dyssynchrony predicts reverse remodeling upon cardiac resynchronization
Am J Physiol Heart Circ Physiol, August 1, 2008; 295(2): H640 - H646.
[Abstract] [Full Text] [PDF]

Home page
 Circ Cardiovasc ImagingHome page
F. W. Prinzen and A. Auricchio
Is echocardiographic assessment of dyssynchrony useful to select candidates for cardiac resynchronization therapy?: Echocardiography Is Not Useful Before Cardiac Resynchronization Therapy if QRS Duration Is Available
Circ Cardiovasc Imaging, July 1, 2008; 1(1): 70 - 78.
[Full Text] [PDF]

Home page
 EuropaceHome page
A. Auricchio and F. W. Prinzen
Update on the pathophysiological basics of cardiac resynchronization therapy
Europace, July 1, 2008; 10(7): 797 - 800.
[Abstract] [Full Text] [PDF]

Home page
 Eur J Heart FailHome page
A. Auricchio and F. F. Faletra
Mechanical dyssynchrony in CRT patients: Still searching for the Holy Grail!
Eur J Heart Fail, March 1, 2008; 10(3): 217 - 219.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (16)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by De Boeck, B. W.L.
Right arrow Articles by Cramer, M. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by De Boeck, B. W.L.
Right arrow Articles by Cramer, M. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?