Population

Ninety-seven patients with ischemic or dilated cardiomyopathy undergoing CRT implantation according to current guidelines at the University Hospital of Rennes were retrospectively included in the study.

All patients were in sinus rhythm and had a good acoustic window, allowing acquisition of bidimensional echocardiography and speckle-tracking echocardiography (STE) with an excellent image quality. At the time of CRT implantation, all patients were receiving optimized medical therapy. CRT response was indicated by a decrease in LV end-systolic volume (ESV) > 15% at follow-up (FU). Clinical data including age, gender, and treatments were collected for each patient. The functional status was assessed by the estimation of the New York Heart Association (NYHA) functional class. An ischemic etiology for LV was claimed in cases with history of myocardial infarction, coronary revascularization, or angiographic evidence of multiple vessel disease or single-vessel disease with ≥75% stenosis of the left main or proximal left anterior descending artery. The study was conducted in accordance with the Good Clinical Practice Guidelines as laid down in the Declaration of Helsinki and reviewed by an independent ethics committee (Regional Ethic Committee validation number: 35RC14-9767). All patients gave their written informed consent.

ECG Data

The 12-lead surface ECG were recorded at 25 and 50 mm/sec during spontaneous rhythm before implantation of the CRT device. The method used for QRS duration analysis has been already reported. LBBB was defined as a QRS duration of ≥120 msec with the following characteristics: QS or rS in lead V1, broad R waves in leads I, aVL, V5, or V6, and absent q waves in leads V5 and V6.

Echocardiography

All patients underwent standard transthoracic echocardiography using a Vivid 7 or Vivid E9 ultrasound system (GE Healthcare, Horten, Norway) equipped with a 3S or M5S 3.5-mHz transducer. Two-dimensional (2D), color Doppler, pulsed wave and continuous wave Doppler data were stored on a dedicated workstation for the offline analysis (EchoPAC, GE Healthcare). LV volumes and function were measured by the biplane method as recommended.

Two-Dimensional Speckle-Tracking Echocardiography

Two-dimensional grayscale images were acquired in the standard apical four-, three- and two-chamber views at a frame rate ≥60 frames/sec. The recordings were processed using an acoustic-tracking dedicated software (EchoPAC version 112.99, Research Release, GE Healthcare), which allowed for an offline semiautomated analysis of speckle-based strain.

To calculate the LV global longitudinal strain (GLS), a line was traced along the LV endocardium’s inner border in each of the three apical views on an end-systolic frame, and a region of interest was automatically identified by the software (EchoPAC, GE Healthcare) between the endocardial and epicardial borders. The sampling of the region of interest was then adjusted to ensure that the wall thickness was incorporated in the analysis, avoiding the pericardium and following myocardial motion, as recommended. The results of segmental and global LV longitudinal strain were then provided by the software. Image quality for the enrolled patients was optimal, and no LV segments were excluded from analysis.

Assessment of Dyssynchrony

Mechanical dyssynchrony was quantified using a multiparametric approach.

• 1.
  

  Atrioventricular delay was calculated as the ratio between LV filling time and the RR interval. Atrioventricular dyssynchrony was considered significant when the duration of LV filling time was <40% of the RR interval.

  
  
• 2.
  

  Interventricular delay was calculated as the time difference between right ventricular to LV ejection. Right ventricular and LV preejection intervals were calculated as the time spans between QRS onset and pulmonary or aortic valve opening determined using Doppler profiles. An interventricular mechanical delay >40 msec was considered an index of interventricular mechanical dyssynchrony.

  
  
• 3.
  

  Intraventricular dyssynchrony was defined by the presence of septal flash (SF) or by the quantification of septolateral or anterior and inferior wall delays measured by STE. A maximal delay >65 msec was indicative of intraventricular dyssynchrony.

Quantification of Myocardial Work

Myocardial work and related indices were estimated using custom software. Myocardial work was estimated as a function of time throughout the cardiac cycle by the combination of LV strain data obtained by STE and a noninvasively estimated LV pressure curve, as previously described by Russell et al. A 17-segment model was used for the estimation of segmental myocardial work.

Estimation of LV Pressure

Peak systolic LV pressure was assumed to be equal to peak arterial pressure measured with a cuff manometer and assumed to be uniform throughout the ventricle. The noninvasive LV pressure curve was then obtained using an empiric, normalized reference curve that was adjusted according to the duration of the isovolumetric and ejection phases defined by the timing of aortic and mitral valve events by echocardiography ( Figure 1 A). Variation in systolic blood pressure as far as variation in valvular events may influence the shape of LV pressure curve ( Figure 1 B). The reliability of this noninvasively estimated LV pressure curve was previously validated in a dog model and in patients with various cardiac disorders.

(A) Example of noninvasive estimation of the LV pressure (LVP) curve. The timing of mitral and aortic valve events is indicated. (B) Examples of variation of the LV pressure curve with measured durations of cardiac phases and with adjustment of peak pressure according to cuff pressure. Left : Basal pressure trace. Middle : Valvular event times are modified ( vertical lines ). Right : Pressure is increased. (C) Representative traces showing PSLs measured in the basal anteroseptal and basal inferoseptal segments in a CRT responder before ( upper ) and after CRT ( lower ). AVC , Aortic valve closure; AVO , aortic valve opening; MVC , mitral valve closure; MVO , mitral valve opening.

Assessment of Myocardial Work by Combination of Pressure and Strain Data

Strain and pressure data were synchronized using the onset R in an electrocardiogram as a common time reference. Myocardial work was then quantified by calculating the rate of segmental shortening by differentiation of the strain curve and multiplying this value with instantaneous LV pressure. This product is a measure of instantaneous power, which was integrated over time to obtain myocardial work as a function of time in systole, which is defined as the time interval from mitral valve closure to mitral valve opening. During the LV ejection period, work performed by the myocardium during segmental elongation represents energy loss, which is defined as WW of that segment. Myocardial work performed during segmental shortening represented CW of that segment. During isovolumetric relaxation, this definition was reversed such that myocardial work during shortening was considered segmental WW and work during lengthening was considered segmental CW. By averaging segmental CW and WW, global CW and WW were estimated for the entire LV.

Figure 1 C presents an example of the PSL traces obtained at the inferobasal and laterobasal segments of a patient with dilated cardiomyopathy and SF before (upper panels) and after (lower panel) successful CRT response. Figure 2 presents an example of GLS, CW, and WW in a CRT responder before and after CRT implantation.

Example of GLS, CW, and WW in a CRT responder before and after CRT implantation. A significant increase in CW and WW is observed after successful CRT. ANT , Anterior; ANT _ SEPT , antero septal; INF , inferior; LAT , lateral; POST , posterior; SEPT , septal.

CRT Delivery

CRT delivery followed a standardized protocol. The right atrial and ventricular leads were positioned conventionally. Preferred localization of the LV lead was a lateral or posterolateral vein. The position was chosen according to the width of QRS, with the goal of obtaining the thinnest one at the end of the procedure. No imaging data were used to identify the site of CRT delivery. After implantation, atrioventricular delay was programmed individually to reach the optimal diastolic filling using the Doppler mitral inflow before discharge, and ventriculoventricular timing was programmed to be simultaneous. After CRT implantation, the LV lead position was confirmed from the chest x-ray.

Statistical Analysis

Statistical analysis was performed using a standard statistical software program (SPSS ver. 20.0, IBM, Chicago, IL).

Continuous variables were expressed by median and interquartile range. Noncontinuous variables were expressed as numbers and percentages. Comparisons between the continuous variables were performed using Mann-Whitney test or the Wilcoxon’s test for paired data. Comparisons between the categorical variables were performed using the χ ² test. The interobserver and intraobserver agreement for CW and WW was assessed on 15 randomly selected subjects by Bland-Altman plot analysis. Interclass coefficients (ICCs) were then calculated as appropriate. A univariate logistic regression analysis was performed to assess the predictive value of clinical features and ECG, echocardiographic, and dyssynchrony parameters with respect to CRT response. Variables with a univariate value <0.05 were inserted in the multivariate analysis (baseline model). A series of nested models were then created by addition of CW and WW alone or in combination. The incremental value of each model was assessed by comparison of χ ² at each step. A receiver operator characteristic (ROC) analysis was used to identify the best cutoff values of CW and WW able to predict CRT response. P < .05 indicated statistical significance.

Population

All clinical, echocardiographic, and myocardial work data from the overall population and based on CRT response are shown in Table 1 . Median LV ejection fraction was 28%; nonischemic etiology for LV dysfunction was identified in 62 patients (64%). After a median FU of 6.1 (5.5-6.9) months, 63 patients (65%) were responders to CRT. CRT+ had greater prevalence of nonischemic cardiomyopathy, interventricular dyssynchrony, and LV dyssynchrony assessed by SF ( Table 1 ).

CW and WW were significantly higher in CRT+. At FU, CRT responders showed a significant improvement in LV ejection fraction (28% [23%-31%] vs 43% [38%-50%]; P < .0001) and GLS (−8.7% [−6.5% to 11.1%] vs −12.3% [−9.30% to 15.35%]; P < .0001) and reduction in LV end-diastolic volume (216 [169-255] vs 155 [117-188] mL; P < .0001) and LV ESV (154 [118-189] vs 88 [60-116] mL; P < .0001). The improvement in LV performance was associated with both an increase in CW (1,093 [783-1,280] vs 1,372 [1,098-1,597] mm Hg%; P < .0001) and a decrease in WW (326 [232-438] vs 199 [122-321] mm Hg%; P < .0001).

Estimation of Myocardial Work and Reproducibility

The estimation of myocardial work was possible in all patients. Interobserver and intraobserver variability is depicted in Figure 3 . The ICCs for the interobserver and intraobserver variability were 0.92 (95% CI, 0.76-0.97; P < .0001) and 0.89 (95% CI, 0.68-0.96; P < .0001) for CW and 0.91 (95% CI, 0.73-0.96; P < .0001) and 0.91 (95% CI, 0.72-0.97; P < .0001) for WW.

Bland-Altman plot analysis for intraobserver and interobserver agreement for CW (A and B) and WW (C and D) . CW _A , WW _A : measures performed by the echocardiographer (A). CW _B , WW _B : measure performed by the echocardiographer (B). CW1, WW ₁ , CW2, WW ₂ : first and second measure performed by the same echocardiographer to test intraobserver agreement.

Predictors of Response to CRT

The factors associated with CRT+ at univariate analysis are shown in Table 2 . LV end-diastolic volume, nonischemic etiology, interventricular dyssynchrony, and SF emerged as significant predictors of CRT+ and were then included in a baseline model (χ ² = 25.11). Three different nested models (models A-C, Table 3 and Figure 4 ) were then used to evaluate the additional information obtained by adding myocardial work parameters to the baseline model. At ROC curve analysis, cutoff values of 1,057 mm Hg% for CW (area under the curve [AUC], 72%; 95% CI [0.62-0.82]; P < .0001) and of 384 mm Hg% for WW (AUC, 0.67; 95% CI [0.57-0.78], P = .005) were associated with CRT+. When these cutoffs were inserted into the multivariate analysis instead of the continuous variables ( Table 3 , model D, and Figure 4 ) they imparted the greatest increase in the model χ ² (46.41, P < .0001). Interestingly, after the inclusion of myocardial work in multivariate analysis, SF remained the only dyssynchrony parameter associated with CRT+.

Stay updated, free articles. Join our Telegram channel

Join