OUP user menu

Mechanical dyssynchrony evaluated by tissue Doppler cross-correlation analysis is associated with long-term survival in patients after cardiac resynchronization therapy

Niels Risum, Eric S. Williams, Michel G. Khouri, Kevin P. Jackson, Niels Thue Olsen, Christian Jons, Katrine S. Storm, Eric J. Velazquez, Joseph Kisslo, Niels Eske Bruun, Peter Sogaard
DOI: http://dx.doi.org/10.1093/eurheartj/ehs035 48-56 First published online: 4 March 2012


Aims Pre-implant assessment of longitudinal mechanical dyssynchrony using cross-correlation analysis (XCA) was tested for association with long-term survival and compared with other tissue Doppler imaging (TDI)-derived indices.

Methods and results In 131 patients referred for cardiac resynchronization therapy (CRT) from two international centres, mechanical dyssynchrony was assessed from TDI velocity curves using time-to-peak opposing wall delay (OWD) ≥80 ms, Yu index ≥32 ms, and the maximal activation delay (AD-max) >35 ms. AD-max was calculated by XCA of the TDI-derived myocardial acceleration curves. Outcome was a composite of all-cause mortality, cardiac transplantation, or implantation of a ventricular assist device (left ventricular assist device) and modelled using the Cox proportional hazards regression. Follow-up was truncated at 1460 days. Dyssynchrony by AD-max was independently associated with improved survival when adjusted for QRS > 150 ms and aetiology {hazard ratio (HR) 0.35 [95% confidence interval (CI) 0.16–0.77], P = 0.01}. Maximal activation delay performed significantly better than Yu index, OWD, and the presence of left bundle branch block (P < 0.05, all, for difference between parameters). In subgroup analysis, patients without dyssynchrony and QRS between 120 and 150 ms showed a particularly poor survival [HR 4.3 (95% CI 1.46–12.59), P < 0.01, compared with the group with dyssynchrony and QRS between 120 and 150 ms].

Conclusion Mechanical dyssynchrony assessed by AD-max was associated with long-term survival after CRT and was significantly better associated compared with other TDI-derived indices. Patients without dyssynchrony and QRS between 120 and 150 ms had a particularly poor prognosis. These results indicate a valuable role for XCA in selection of CRT candidates.

  • Cardiac resynchronization therapy
  • Dyssynchrony
  • Cross-correlation analysis
  • Long-term outcome


Cardiac resynchronization therapy (CRT) has been demonstrated to reduce morbidity and mortality in randomized clinical trials of patients with left ventricular ejection fraction (LVEF) ≤35%, QRS duration ≥120 ms, and symptomatic heart failure despite medical therapy.14 However, more than a third of all patients receiving a CRT device are non-responders, indicating that these current criteria for selecting CRT candidates may be suboptimal. Although non-response is multifactorial, the lack of pre-implant mechanical dyssynchrony is likely to play a key role5 as subgroup analyses from the major clinical trials suggest that CRT is less effective in patients without left bundle branch block (LBBB) QRS morphology6 and patients with QRS duration <150 ms.1,3,4 Lower response rates among these subgroups may reflect an actual lack of baseline LV dyssynchrony6 and supports a potential role for direct assessment of mechanical dyssynchrony to select patients for CRT.

A virtual bouquet of methods has been proposed to evaluate mechanical dyssynchrony as a means of predicting CRT response. Among these, measurement of time-to-peak differences in mechanical events is most common.710 Pre-implant dyssynchrony assessment based on this approach has been shown in some studies to predict long-term survival after CRT.5,7,1113 One multicentre study, PROSPECT, however, found these indices to be too unreliably measured to be clinically useful and that further study was warranted.14 Other methods have emerged to assess the opposing wall motion characteristic of intraventricular dyssynchrony.1517 Cross-correlation analysis (XCA) of myocardial systolic acceleration is one such method that predicts LV remodelling after CRT.17

XCA shows potential for reducing the subjectivity of wave form analysis and its mathematical approach offers a comprehensive method for mechanical dyssynchrony estimation, not dependent on single measurements of the complex wave forms of mechanical cardiac movement. The aims of this study were: (i) to assess the impact of echocardiographic dyssynchrony measured by XCA on long-term outcomes in CRT recipients, (ii) to compare this method to other commonly used parameters of longitudinal dyssynchrony, and (iii) to study this method with respect to QRS duration.


Study population

We retrospectively studied all echocardiograms and medical records of patients who received an echocardiographic dyssynchrony study [tissue Doppler imaging (TDI)] prior to CRT implantation at two international centres between August 2004 and April 2007. These subjects comprised a consecutive cohort of patients with LVEF ≤ 35%, QRS duration ≥ 120 ms, and New York Heart Association II–IV functional class, despite optimal pharmacological therapy. Overall, 131 patients were identified (71 patients from Duke University Medical Center, Durham, NC, USA, between 2004 and 2006 and 60 patients from Gentofte University Hospital, Copenhagen, Denmark, between 2006 and 2007). In none of the patients was the decision to implant a CRT device based on the dyssynchrony study. In addition to demographic and clinical data, baseline data collection included electrocardiography, echocardiography, and routine laboratory work. Renal function was assessed by estimated glomerular filtration rate (eGFR), calculated from the abbreviated Modification of Diet in Renal Disease formula study equation.18 Subjects were excluded if they had significant primary valve disease, atrial fibrillation, or acute coronary syndrome or revascularization within 3 months of the baseline echocardiography.

All patients were implanted with a CRT device with defibrillator capacity according to the standard clinical practice with one lead in the high right atrium, a right ventricular apical or septal lead, and an LV lead positioned through the coronary sinus in an epicardial vein targeting posterolateral or lateral branches.

The study protocol was approved by the Institutional Review Board at both centres and complied with the Declaration of Helsinki.


All echocardiographic studies were performed with a standard imaging system and software (VIVID 7, GE-Vingmed, Horton, Norway) by experienced sonographers including obtainment of colour TDI cineloops of one representative cardiac cycle in each of the three standard apical views. The sector width and depth were optimized to include the LV only, and colour scale was adjusted to avoid aliasing. All TDI studies were performed at a frame rate between 95 and 150 s−1.

Off-line analysis was performed using EchoPac PC (version BT09, GE Vingmed Ultrasound). The LV end-systolic volume, LV end-diastolic volume, and LVEF were assessed using the biplane Simpson's method. Diastolic function was graded 0–4 according to guidelines.19

Cross-correlation analysis of myocardial acceleration

The use of XCA has previously been explained in details17 and is shown in Figure 1.

Figure 1

From tissue velocity traces to cross-correlation analysis of acceleration. Upper panel: the generation of tissue Doppler imaging longitudinal velocity curves from left ventricular apical four-chamber basal sites. Blue curve shows the basal septal wall and red curve the basal lateral wall. Lower panel: the same patient. Differentiation of tissue Doppler velocity traces (left column) generates the acceleration traces (middle column). Cross-correlation coefficient spectrum (right column) is calculated from the systolic periods (between dashed lines) of the acceleration curves. This example shows a 68-year-old man, who did not respond to cardiac resynchronization therapy. The patient had ischaemic dilated cardiomyopathy, left bundle branch block, New York Heart Association class III and left ventricular ejection fraction of 28%. Time-to-peak velocity analysis showed a significantly delayed lateral wall (158 ms between vertical arrows). In contrast, acceleration analysis yielded a high XXC0 (0.74) and a low activation delay of 14 ms (horizontal arrows). Thus, the patient did not have significant dyssynchrony.

Briefly, TDI traces of myocardial velocity was obtained off-line by placing regions of interest (ROIs) (7 × 15 mm) at the basal LV segments of opposing walls in the TDI apical four-chamber, two-chamber, and long-axis views. The ROI's were carefully adjusted to find the most reproducible velocity curves. The velocity traces were then exported as text files (together with the electrocardiogram) into a customized spreadsheet (Microsoft Excel 2003; Microsoft Corp., Redmond, WA, USA) for XCA.

The time period of interest between QRS onset and aortic valve closure was defined and the velocity data were converted to acceleration by temporal differentiation. To reduce noise, three-point filtering was used. Acceleration traces from opposing walls were then automatically compared by the spreadsheet using a cross-correlation algorithm according to the following formula:Embedded Image where XCCd is the cross-correlation coefficient at time shift d, xi the acceleration trace from LV basal segment, yi the acceleration trace from opposing LV basal segment, and Embedded Image and Embedded Image represent the mean values.

The XCC was a measure of the degree of association between the acceleration traces, i.e. it investigated the similarity of the direction and magnitude of the acceleration in the two opposing walls. To produce the XCC spectrum (Figure 1, lower panel, right), traces were automatically time-shifted frame by frame, calculating an XCC for each time shift, and the time shift that resulted in the highest XCC was the activation delay (AD). The analysis was performed in each of the three apical views to obtain the maximal absolute AD (AD-max) in each patient. Significant dyssynchrony by AD-max was defined as AD-max > 35 ms as reported previously.17

Acceleration data were used because cross-correlation requires stationary means of the compared signals and translational or rotational motion of the heart can cause a drift in the velocity signal during systole which may affect the cross-correlation analysis. This can be compensated for by differentiating the velocity traces to obtain myocardial acceleration. Although acceleration traces are noisier, and additional filtering was necessary, cross-correlation analysis, and in particular the AD, is noise-insensitive because it is based on the overall patterns of many acceleration values. A customized spreadsheet for cross-correlation analysis is provided in Supplementary material online.

Conventional time-to-peak analysis of myocardial velocity

Time-to-peak systolic velocity was measured from the onset of the QRS complex, excluding isovolumic periods (Figure 1). Aortic valve opening and closure were defined by using TDI curved m-mode through the anterior mitral leaflet.20 Dyssynchrony was assessed by opposing wall delay (OWD), defined as the maximal time difference in peak velocity at basal and mid-segments in opposing walls for each view.5 The Yu index was calculated as the 12-site time-to-peak velocity standard deviation (SD) in basal and mid-segments.10 Significant longitudinal dyssynchrony by TDI was pre-defined as the maximal OWD in one view ≥80 ms5 or Yu index ≥32 ms.5,10 The acquisition of TDI traces and all dyssynchrony analyses were performed by an interpreter blinded to information about the clinical status and outcome of the patients.

Long-term outcome and subgroup analyses

The primary outcome was a composite of all-cause mortality, cardiac transplantation, or implantation of a ventricular assist device [left ventricular assist device (LVAD)]. Vital status for all subjects was ascertained through chart review, the United States Social Security Death Index, and the Danish civil registration register, respectively, at the time of study analysis in November 2011. Other clinical data were ascertained by reviewing cardiology and electrophysiology clinic notes. Pre-defined subgroup analyses were planned in subjects with QRS duration between 120–150 and >150 ms.

Statistical analysis

Relevant variables were tested for normality using visual inspection of histogram plots and are presented as mean ± SD. Continuous variables were compared using Student's t-test. Proportional differences were tested using χ2 statistics or Fisher's exact test where appropriate. For all survival analyses, follow-up was truncated at a maximum of 4 years (1460 days). The cumulative probability of the endpoint was illustrated using the Kaplan–Meier method with significance testing using log-rank statistics. Univariate and multivariable predictors of event-free survival after CRT device implantation were assessed in Cox's proportional hazards models. Proportional hazards assumptions were verified graphically. We included covariates previously shown to predict mortality from heart failure including age, gender, LBBB, QRS duration >150 ms, ischaemic heart disease, renal function, and LVEF.1,6,21,22 Age and LVEF were analysed as dichotomous variables using the median value for each variable for the entire study population. Chronic kidney disease was defined by the clinically accepted definition of eGFR < 60 mL/min/1.73 m2.22 Candidate variables with P-values of <0.1 in univariate analysis were included in the multivariable model using backward selection to test the independent association between outcome and each dyssynchrony index (OWD ≥ 80 ms, Yu index ≥32 ms, and AD-max > 35 ms).

For comparison between indices, the strength of association for each dyssynchrony index was compared using −2 log-likelihood statistics. Receiver operating characteristic (ROC) curve analysis was performed for each dyssynchrony index with the use of a non-parametric estimate of the area under the curve (AUC) and c-statistics with 95% confidence interval (CI) were performed for each multivariable model.23 Furthermore, the ability of AD-max > 35 ms to reclassify patient risk when added to a multivariable model with either Yu index ≥32 ms or OWD ≥ 80 ms was evaluated by assessment of the net reclassification improvement (NRI).24 Patients were initially classified as at low or high risk of an event if their predicted risk was <10 or ≥10%, respectively.3 Patients were then reclassified (or unchanged) into a different category with the addition of AD-max > 35 ms.

A two-tailed P-value of <0.05 was considered significant in the final models. Intra- and interobserver variability was determined from 20 randomly selected studies and assessed as the mean difference ± the coefficient of variation (SD/mean). All statistical analyses were performed using SAS for Windows version 9.1.3 (SAS institute, Cary, NC, USA).


Feasibility and variability of dyssynchrony analysis

Of the 131 patients included, assessment of AD-max was possible in 121 patients (92%) due to inadequate image quality in 7 patients and atrial fibrillation in 3 patients. Five patients had inadequate image quality at the mid-ventricular level; thus, evaluation of the Yu index and OWD was possible in 116 patients (89%). The intra- and interobserver reproducibility was 1 ms ± 6% and 3 ms ± 7% for Yu index; 2 ms ± 6% and 3 ms ± 7% for OWD; and −2 ms ± 4% and −3 ms ± 4% for AD-max, respectively.

Baseline characteristics

The baseline clinical data in relation to AD-max for the study population are shown in Table 1. Eighty patients (66%) had LBBB, 25 patients (20%) had non-specific interventricular conduction delay and 16 patients (13%) had right bundle branch block. The prevalence of dyssynchrony varied according to the index used, with 53% of the patients having dyssynchrony by AD-max, 86% by Yu index, and 50% by OWD. Baseline characteristics among patients with and without dyssynchrony by AD-max differed significantly in some important respects. Dyssynchrony was primarily found in patients with LBBB and less often in non-LBBB patients (69 vs. 22%, P < 0.001). Furthermore, a larger percentage of women were observed in the group with dyssynchrony than in the group without (33 vs. 9%, P < 0.05).

View this table:
Table 1

Baseline characteristics and relation to the maximal activation delay

≤35 ms>35 ms
Age, years65.7 ± 1066.9 ± 964.8 ± 11
Female, n (%)26 (22)5 (9)21 (33)*
NYHA2.9 ± 0.43 ± 0.42.9 ± 0.4
Diastolic grade2.2 ± 0.42.3 ± 1.12.1 ± 1.2
QRS, ms155 ± 22150 ± 23159 ± 18
LBBB, n (%)80 (66)25 (43)55 (86)*
Non-LBBB, n (%)41 (34)32 (56)9 (14)*
Ischaemic aetiology, %72 (60)39 (68)33 (52)
eGFR, unit56 ± 2054 ± 1958 ± 20
LVEF, %22.5 ± 822 ± 823 ± 8
LVESV, mL162 ± 62166 ± 70159 ± 63
LVEDV, mL205 ± 70214 ± 72202 ± 99
β-Blockers, n (%)105 (87)49 (86)56 (87)
ACE/AII blocker, n (%)109 (90)52 (91)57 (89)
Spironolactone, n (%)54 (45)28 (50)26 (41)
Diuretics, n (%)87 (72)45 (79)42 (65)
  • NYHA, New York Heart Association functional class; LBBB, left bundle branch block; eGFR, estimated glomerular filtration rate; LVEF, left ventricular ejection fraction; LVESV, left ventricular end-systolic volume; LVEDV, left ventricular end-diastolic volume. Values represent mean ± SD.

  • *P < 0.05 vs. patients with AD-max ≤ 35 ms.

Outcome and relation with dyssynchrony

One hundred and seventeen (97%) patients had maximum 4-year follow-up. The remaining four patients with incomplete follow-up were followed 1, 178, 902, and 1263 days, respectively. The patient with 1-day follow-up was not included in the final models. During the follow-up period, 31 patients (26%) died, one patient (0.8%) received an LVAD and none of the patients had a heart transplant (HTX). Nine patients (7%) died within 6 months of CRT implantation.

Dyssynchrony by AD-max was associated with a significantly longer event-free survival compared with patients without dyssynchrony (Figure 2). Assessed by AD-max, only 14% of the patients with dyssynchrony died compared with 40% in the group without dyssynchrony during follow-up (P = 0.001). The Yu index and OWD only showed borderline significant association with outcome (P = 0.09 and 0.06, respectively) as shown in Figures 3 and 4. Among covariates, there was a significant association between event-free survival and LBBB morphology, non-ischaemic aetiology, and QRS duration >150 ms (Table 2). These covariates were included in the multivariable model with each of the dyssynchrony indices added separately. However, due to a high degree of co-linearity for LBBB with all dyssynchrony indices, this parameter was not associated with outcome in the multivariable model and excluded from subsequent models. Of note, AD-max showed a significantly higher association with outcome compared with LBBB in the multivariable model (P = 0.04, for difference between parameters).

View this table:
Table 2

Univariate and multivariate risk analyses of mortality after cardiac resynchronization therapy

HR95% CIP-value
Univariate analysis
 Age >651.40.68–2.800.35
 Ischaemic aetiology2.361.06–5.270.03*
 QRS ≤ 150 ms2.010.99–4.010.05
 LVEF < 22%1.420.71–2.840.32
 eGFR < 601.640.79–3.40.18
 AD-max > 35 ms0.260.11–0.580.001*
 OWD ≥ 80 ms0.490.23–1.040.06
 Yu index ≥32 ms0.460.19–1.130.09
Multivariate analysis: each index added individually
 AD-max > 35 ms0.350.16–0.770.01*
 Yu index ≥32ms0.49020–1.200.12
 OWD ≥ 80 ms0.530.25–1.110.09
  • Univariate and multivariate analyses of relevant baseline characteristics. For multivariate analysis, each dyssynchrony parameter was added individually to the model including QRS > 150 ms and aetiology. AD-max was the only parameter of dyssynchrony independently associated with response. CI, confidence interval; other abbreviations as in Table 1.

  • *P < 0.05.

Figure 2

The Kaplan–Meier curves of freedom from death, left ventricular assist device or heart transplant after cardiac resynchronization therapy in relation to the maximal activation delay. Patients with dyssynchrony had a more favourable outcome than those without. *P = 0.001.

Figure 3

The Kaplan–Meier curves of freedom from death, left ventricular assist device or heart transplant after cardiac resynchronization therapy in relation to the Yu index. Difference in outcome between groups was not significant.

Figure 4

The Kaplan–Meier curves of freedom from death, left ventricular assist device or heart transplant after cardiac resynchronization therapy in relation to opposing wall delay. Difference in outcome between groups was not significant.

In multivariate analysis, the presence of dyssynchrony by AD-max remained independently associated with event-free survival [hazard ratio (HR) 0.35 (95% CI 0.16–0.77), P = 0.01]. However, OWD and Yu index were not independently associated with outcome after adjustment for other clinical variables, despite borderline significance (Table 2).

When comparing indices, dyssynchrony evaluated by AD-max > 35 ms had a significantly higher association with outcome compared with Yu index ≥32 ms and OWD ≥ 80 ms in the multivariable models (P = 0.01, both, for difference between parameters).

The ROC curves for the three dyssynchrony indices are shown in Figure 5. Sensitivity and specificity for survival free from HTX and LVAD was 63 and 73% for AD-max > 35 ms, 89 and 20% for Yu index ≥32 ms, and 55 and 63% for OWD ≥ 80 ms. c-statistics for the corresponding multivariable models were 0.77 (95% CI 0.67–0.86), 0.70 (95% CI 0.6–0.82), and 0.65 (95% CI 0.53–0.77). When comparing c-statistics for AD-max to the other indices, no statistically significant differences were found (P = 0.3 compared with Yu-index and P = 0.07 compared with OWD).

Figure 5

Receiver operating characteristics curve for the associations of baseline dyssynchrony by AD-max, Yu index, and opposing wall delay with survival free from heart transplant or left ventricular assist device after cardiac resynchronization therapy implantation. AUC, area under the curve.

Adding AD-max > 35 ms to a 4-year 10% risk model with QRS > 150 ms, ischaemic heart disease, and Yu index ≥32 ms caused significant NRIs. Twenty-one patients were appropriately moved to the lower-risk category and only one patient inappropriately to the high-risk group, while two patients with an event were appropriately reclassified to the high-risk category (NRI = 0.3, P < 0.001). Similarly, in the model with OWD ≥ 80 ms, adding AD-max improved the model by a desirable downward risk reclassification for 26 non-event patients and only 1 patient who had an event was inappropriately reclassified to the low-risk category (NRI = 0.27, P < 0.001). Improvements in reclassification were driven by significant improvements in specificity for event-free survival when AD-max was added to the models (P < 0.02, for both models).

Finally, we tested the potential relationship of QRS duration and dyssynchrony by AD-max in subgroup analyses (Figure 6). Patients without dyssynchrony and QRS between 120 and 150 ms showed a particularly poor survival when compared with groups with dyssynchrony [HR 4.29 (95% CI 1.46–12.59) compared with AD-max > 35 ms and QRS between 120 and 150 ms, P = 0.008; HR 4.03 (95% CI 1.91–8.51) compared with AD-max > 35 ms and QRS > 150 ms, P < 0.001]. There was a borderline significant interaction between QRS duration and AD-max (P = 0.07).

Figure 6

The Kaplan–Meier curves showing the probability of freedom from death, left ventricular assist device or heart transplant when patients were subdivided by baseline QRS duration and mechanical dyssynchrony evaluated by the maximal activation delay. Note the particularly poor prognosis of patients with QRS duration between 120 and 150 ms without dyssynchrony (*P < 0.01 when compared with AD-max > 35 ms and QRS between 120–150 ms).


Benefits from cardiac resynchronization therapy and its relation to mechanical dyssynchrony

The primary beneficial mechanism of CRT is believed to be correction of mechanical dyssynchrony resulting in improved coordination of ventricular contraction. In support of this hypothesis, mechanical dyssynchrony before CRT implantation has been shown to correlate with improvements in quality of life, in 6 min walking distance, in LV remodelling, and in improvements in LVEF.810,15,17,25 In addition, previous studies have shown the presence of echocardiographic dyssynchrony prior to CRT to be predictive of long-term survival considered independently5,7,1113 or in combination with clinical markers.26

Despite its potential impact in selecting candidates for therapy, assessment of mechanical dyssynchrony before CRT implantation has not entered clinical guidelines, and echocardiographic methods used to assess mechanical dyssynchrony have been subject to some criticism, especially after the PROSPECT study showed this assessment to be highly variable between observers.14

The current study adds value to this field by (i) confirming the strong association between mechanical dyssynchrony and favourable outcome after CRT and (ii) demonstrating that XCA of myocardial acceleration has a stronger association with outcome than other tissue Doppler-based methods. Thus, this new, conceptually different, and more robust method for echocardiographic dyssynchrony estimation may potentially overcome the objections raised by the PROSPECT study.

Cross-correlation analysis for assessment of mechanical dyssynchrony

We have recently validated XCA and demonstrated its ability to better discriminate wall motion abnormalities caused by other diseases from those caused by dyssynchrony and to predict LV reverse remodelling after CRT.17 The current study demonstrates the association of XCA-derived AD-max with long-term survival after CRT. In addition, our results show that AD-max has a significantly stronger association with long-term outcome than other TDI-derived parameters of longitudinal dyssynchrony. Net reclassification improvement analysis suggests that adding AD-max to a predictive model including QRS > 150 ms, ischaemic heart disease, and either Yu index or OWD will result in a considerable improvement in risk classification, especially by identifying patients with low risk of death, HTX or LVAD if implanted with a CRT device. Of course, this finding must be cautiously interpreted as the risk categories are based on a somewhat arbitrary choice of categories.

Most commonly, longitudinal dyssynchrony is quantified by various indices of time-to-peak differences in opposing wall velocities by TDI.5,7,11 Recently, Gorcsan et al.5 showed that the Yu index ≥32 ms and OWD ≥ 80 ms in all three apical views were associated with long-term event-free survival, although only the Yu index was found to be independently associated with outcome in multivariable analysis. In the current study, the Yu index and OWD only showed borderline significant association with outcome. While lack of significant association for these indices may partly be ascribed to the sample size, more importantly, applying XCA to the same velocity data significantly improved the association of mechanical dyssynchrony to outcome.

The PROSPECT study failed to show any use of echocardiographic measurements in the evaluation of dyssynchrony and caused concerns about the feasibility and reproducibility of TDI-derived time-to-peak parameters in particular.14 Thus, although PROSPECT had several well-acknowledged limitations such as low image quality and lack of standardization,5 it also raised the issue of relying on single-point measurements when assessing dyssynchrony. For instance, while time-to-peak velocities may reflect dyssynchronous activation to a certain extent, significant time-to-peak differences can be observed merely due to heterogeneity of the failing heart.17 Presumably, such mechanical disturbances will not be amenable to CRT.27 XCA uses instantaneous information from the entire period of myocardial activation which improves the ability to differentiate true inter-wall timing differences from other regional differences in wall performance, and it is particularly sensitive to periods of opposing wall motion which is characteristic of intraventricular dyssynchrony.15,16,28,29 Due to its automatic calculation, it also reduces observer variability compared with the often ambiguous measurement of time-to-peak.

Potential clinical use of dyssynchrony assessment

Evaluation of mechanical dyssynchrony may have only a limited clinical role to play in patients with QRS > 150 ms, as these patients have a large a priori likelihood of response,1,3,4 as evidenced in this study. In patients with intermediate QRS duration, the picture is different. A recent meta-analysis including the five largest CRT trials showed that patients with QRS durations between 120 and 149 ms did not benefit from CRT.30 Our results indicate a particularly poor prognosis for CRT-treated patients without dyssynchrony in the narrower-QRS group. These results are in line with recent findings reported by Gorcsan et al.5 using the Yu index and two-dimensional (2D) radial strain. Thus, it appears that accurate assessment of mechanical dyssynchrony may be especially valuable in selection of patients with QRS duration between 120 and 150 ms. Future studies on the clinical value of mechanical dyssynchrony assessment are warranted specifically focusing on this subgroup of patients.


This study was not randomized and there was no comparison to patients who did not undergo CRT. Consequently, in the absence of a control group, the direct treatment effect of CRT cannot be ascertained. Thus, the potential negative impact of CRT in patients without dyssynchrony was not investigated and remains unknown. While intriguing, our findings should not yet be used to supplant established clinical guidelines or criteria for CRT therapy as prospective and randomized studies are warranted.

We were not able to demonstrate a statistically significant independent association of OWD and Yu index with outcome and these findings are limited by the relatively small sample size in this study. However, all dyssynchrony analyses were performed from the same data set emphasizing the value of XCA.

Other studies have demonstrated the relationship of mechanical dyssynchrony with long-term outcome assessing time-to-peak radial function.5,12,13 Assessment of radial function using speckle-tracking analysis may have a better correlation than existing indices of longitudinal function.5,13 Comparison to 2D-radial strain was not possible in this study, as images optimized for 2D-strain analysis were not consistently acquired.

Finally, we did not assess the LV lead position or scar tissue burden, factors known to impact CRT response.21,31


Mechanical dyssynchrony assessed by XCA of echocardiographic tissue Doppler-derived myocardial motion is independently associated with long-term survival after CRT. Dyssynchrony by this modality was significantly stronger associated with outcome than other tissue Doppler-derived indices of longitudinal mechanical dyssynchrony. Our findings suggest an important role for improved dyssynchrony assessment prior to CRT implantation.

Supplementary material

Supplementary material is available at European Heart Journal online.

Conflict of interest: J.K. is a speaker for Phillips and GE and gives advice on product development.


View Abstract