The analysis of left ventricular (LV) mechanics provides novel insights into the effects of cardiac resynchronization therapy (CRT) on LV performance. Currently, advances in speckle-tracking echocardiographic analysis have permitted the characterization of subendocardial and subepicardial LV twist. The aim of this study was to investigate the role of the acute changes in subendocardial and subepicardial LV twist for the prediction of midterm beneficial effects of CRT. A total of 84 patients with heart failure scheduled for CRT were recruited. All patients underwent echocardiography before and <48 hours after CRT implantation and at 6-month follow-up. The assessment of LV volumes, ejection fractions, and mechanical dyssynchrony (systolic dyssynchrony index) was performed with real-time 3-dimensional echocardiography. The assessment of subendocardial and subepicardial LV twist was performed with 2-dimensional speckle-tracking echocardiography. A favorable outcome was defined as the occurrence of a reduction ≥15% in LV end-systolic volume associated with an improvement of ≥1 New York Heart Association functional class at 6-month follow-up. At 6-month follow-up, 53% of the patients showed favorable outcomes. Ischemic cause of heart failure, baseline systolic dyssynchrony index, immediate improvement in the LV ejection fraction, immediate improvement in systolic dyssynchrony index, and immediate improvement in subendocardial and subepicardial LV twist were significantly related to favorable outcomes. However, in multivariate logistic regression analysis, only the immediate improvement of subepicardial LV twist was independently related to favorable outcomes (odds ratio 2.31, 95% confidence interval 1.29 to 4.15, p = 0.005). Furthermore, the immediate improvement of subepicardial LV twist had incremental value over established parameters. In conclusion, the immediate improvement of subepicardial LV twist (but not subendocardial LV twist) is independently related to favorable outcomes after CRT.
Recent advances in speckle-tracking echocardiography analysis have permitted the characterization of subendocardial and subepicardial left ventricular (LV) twist. The aim of the present study was to investigate the changes induced by cardiac resynchronization therapy (CRT) on the rotational mechanics detected with speckle-tracking echocardiography. Specifically, the role of the acute changes in subendocardial and subepicardial LV twist for the prediction of midterm beneficial effects of CRT (LV reverse remodeling associated with clinical improvement at 6-month follow-up) was explored over the classical parameters, including mechanical LV dyssynchrony and the LV ejection fraction (LVEF).
Methods
A total of 106 consecutive patients with heart failure scheduled for CRT were included. According to current guidelines, the inclusion criteria were New York Heart Association functional class III or IV, sinus rhythm, LVEF ≤35%, and QRS duration ≥120 ms. The cause of heart failure was considered ischemic in the presence of significant coronary artery disease (>50% stenosis in ≥1 major epicardial coronary artery) on coronary angiography and/or a history of myocardial infarction or revascularization.
All patients underwent complete baseline clinical evaluations, 12-lead surface electrocardiography, and transthoracic echocardiography before and <48 hours after CRT device implantation. Global measures of LV performance were evaluated using real-time 3-dimensional echocardiography and 2-dimensional speckle tracking. The assessment of LV volumes, LVEFs, and mechanical dyssynchrony was performed using real-time 3-dimensional echocardiography. The assessment of subendocardial and subepicardial LV twist was performed using 2-dimensional speckle tracking. In addition, the clinical and echocardiographic evaluations were repeated 6 months after CRT implantation. A favorable outcome was defined as the occurrence of LV reverse remodeling (a reduction ≥15% in LV end-systolic volume) associated with clinical improvement (≥1 New York Heart Association functional class) at 6-month follow-up. Finally, the variables related to favorable outcomes were investigated, and the role of the newest rotational parameters over the established parameters was explored.
Images were obtained with patients in the left lateral decubitus position using a commercially available system (iE33; Philips Medical Systems, Bothell, Washington) equipped with an X3 fully sampled matrix transducer. Apical full-volume data sets were obtained at frame rates of 20 to 35 frames/s, and quantitative analysis was performed off-line using a semiautomated contour-tracing algorithm (Q-Lab version 6.0; Philips Medical Systems), as previously described. LV volumes and LVEFs were measured. In addition, the systolic dyssynchrony index (SDI) was obtained as marker of global LV dyssynchrony, as previously reported.
Two-dimensional grayscale harmonic images were obtained in the left lateral decubitus position using a commercially available ultrasound system (iE33) equipped with a broadband S5-1 transducer. Parasternal short-axis images were acquired at 2 different levels: (1) the basal level, identified by the mitral valve, and (2) the apical level, defined as the smallest cavity achievable distally to the papillary muscles (just proximal to the level with end-systolic luminal obliteration), moving the probe down and slightly laterally. Patients without appropriate apical levels were excluded from the study. Frame rates ranged from 55 to 90 frame/s, and 3 cardiac cycles for each parasternal short-axis level were stored in cine loop format for off-line analysis (Q-Lab version 6.0).
In the present study, 2-dimensional speckle-tracking analysis (Q-Lab version 6.0) was performed by placing manually several small kernel regions in the subendocardial and subepicardial border on an end-diastolic frame. Then, the software tracked the 2 borders frame by frame, and tracking could be adjusted manually if needed.
The 2-dimensional speckle-tracking software calculates LV rotation from the apical and basal short-axis images as the average angular displacement of the 6 standard segments referring to the ventricular centroid, frame by frame. Counterclockwise rotation was marked as positive values and clockwise rotation as negative values when viewed from the LV apex. LV twist was defined as the net difference (in degrees) of apical and basal rotation at isochronal time points. For the calculation of LV twist, averaged apical and basal rotation data were exported to a spreadsheet program (Excel 2003; Microsoft Corporation, Redmond, Washington). The following measurements were derived for subendocardial and subepicardial layers: peak apical and basal rotation and peak LV twist.
All patients received biventricular pacemakers with cardioverter-defibrillator function (Contak Renewal 4RF, Boston Scientific Corporation, St. Paul, Minnesota; InSync Sentry, Medtronic, Inc., Minneapolis, Minnesota; or Lumax 340 HF-T, Biotronik GmbH, Berlin, Germany). The right atrial and ventricular leads were positioned conventionally. All LV leads were implanted transvenously and positioned preferably in a lateral or posterolateral vein. A coronary sinus venogram was obtained using a balloon catheter, followed by the insertion of the LV pacing lead. An 8Fr guide catheter was used to place the LV lead (Easytrak, Boston Scientific Corporation; Attain-SD, Medtronic, Inc.; or Corox OTW, Biotronik GmbH) in the coronary sinus.
All continuous variables are presented as mean ± SD. Categorical data are presented as numbers and percentages. Unpaired Student’s t tests were used to compare parameters at baseline, immediately after CRT, and at 6-month follow-up between patients with and those without favorable outcomes. Chi-square tests were used to compare categorical variables between patients with and without favorable outcomes. Paired Student’s t tests were used to compare baseline and 6-month follow-up data in each group of patients. To identify variables related to favorable outcomes at 6-month follow-up, univariate and multivariate logistic regression analysis were performed, including the clinical and echocardiographic characteristics of the patients at baseline and immediately after CRT. Only significant (p <0.05) univariate factors were entered as covariates in the multivariate analysis, using a backward selection model. Finally, the incremental value of the newest rotational parameters (subendocardial and subepicardial LV twist) over other variables was assessed by calculating the global chi-square value for each model. All statistical tests were 2 sided and p values <0.05 were considered significant. SPSS version 16.0 (SPSS, Inc., Chicago, Illinois) was used for statistical analysis.
Results
A total of 22 of 106 patients (21%) were excluded because the image quality did not allow reliable analysis. Therefore, the overall patient population consisted of 84 patients. Of the 84 patients with heart failure enrolled, 4 did not complete the 6-month follow-up; 1 patient died of worsening heart failure, 1 had LV pacing switched off because of intolerable phrenic stimulation, and 2 were lost to follow-up.
Baseline characteristics of the overall patient population are listed in Table 1 . The mean age was 65 ± 9 years, and all patients had dilated left ventricles with poor LV function (mean LVEF 26 ± 5%). In addition, the LV rotational parameters were severely reduced in the subendocardial and subepicardial layers (peak subendocardial and subepicardial LV twist were 4.2 ± 2.9° and 2.3 ± 1.8°, respectively).
| Variable | Value |
|---|---|
| Age (years) | 65 ± 9 |
| Men/women | 55/29 |
| Medication | |
| Angiotensin-converting enzyme inhibitors | 76 (91%) |
| β blockers | 71 (85%) |
| Diuretics and/or spironolactone | 71 (85%) |
| Cause of heart failure | |
| Ischemic | 44 (52%) |
| Nonischemic | 40 (48%) |
| NYHA functional class III/IV | 81/3 |
| QRS duration (ms) | 151 ± 29 |
| Left ventricular end-diastolic volume (ml) | 200 ± 61 |
| Left ventricular lead positioned in posterolateral vein | 80 (95%) |
| Left ventricular end-systolic volume (ml) | 147 ± 50 |
| Left ventricular ejection fraction (%) | 26 ± 5 |
| Systolic dyssynchrony index (%) | 7.3 ± 2.1 |
| Peak subendocardial apical rotation (°) | 2.5 ± 2.1 |
| Peak subendocardial basal rotation (°) | −2.5 ± 2.1 |
| Peak subendocardial left ventricular twist (°) | 4.2 ± 2.9 |
| Peak subepicardial apical (°) | 1.5 ± 1.4 |
| Peak subepicardial basal rotation (°) | −1.7 ± 1.4 |
| Peak subepicardial left ventricular twist (°) | 2.3 ± 1.8 |
At 6-month follow-up, 42 of 80 patients (53%) showed favorable outcomes (LV reverse remodeling associated with clinical improvement at 6-month follow-up).
The baseline parameters of the patients with and without favorable outcomes (LV reverse remodeling associated with clinical improvement at 6-month follow-up) are listed in Table 2 . Patients with favorable outcomes less frequently had ischemic causes of heart failure (p = 0.025). The 2 groups of patients had comparable LV volumes and LVEFs. However, the baseline SDI was significantly larger in the group of patients with favorable outcomes (8.1 ± 2.3% vs 6.4 ± 1.4%, p <0.001). In addition, no differences in baseline subendocardial LV twist (4.0 ± 3.1° vs 4.5 ± 2.7°, p = 0.45) or subepicardial LV twist (1.9 ± 1.8° vs 2.6 ± 1.8°, p = 0.10) were observed.
| Variable | Patients With Favorable Outcomes | Patients Without Favorable Outcomes | p Value |
|---|---|---|---|
| (n = 42) | (n = 38) | ||
| Ischemic cause of heart failure | 16 (38%) | 24 (63%) | 0.025 |
| QRS duration (ms) | 152 ± 27 | 151 ± 31 | 0.87 |
| NYHA functional class I/II/III/IV | |||
| Baseline | 0/0/41/1 | 0/0/36/2 | 0.50 |
| 6-month follow-up | 10/31/1/0 ⁎ | 2/8/26/2 ⁎ | <0.001 |
| LV end-systolic volume (ml) | |||
| Baseline | 150 ± 47 | 143 ± 54 | 0.50 |
| 6-month follow-up | 107 ± 39 ⁎ | 149 ± 54 | <0.001 |
| LV ejection fraction (%) | |||
| Baseline | 26 ± 6 | 26 ± 5 | 0.97 |
| 6-month follow-up | 38 ± 7 ⁎ | 27 ± 7 | <0.001 |
| SDI (%) | |||
| Baseline | 8.1 ± 2.3 | 6.4 ± 1.4 | <0.001 |
| 6-month follow-up | 5.1 ± 1.8 ⁎ | 8.0 ± 3.8 | <0.001 |
| Peak subendocardial LV twist (°) | |||
| Baseline | 4.0 ± 3.1 | 4.5 ± 2.7 | 0.45 |
| 6-month follow-up | 5.7 ± 2.9 ⁎ | 3.8 ± 2.8 | 0.004 |
| Peak subepicardial LV twist (°) | |||
| Baseline | 1.9 ± 1.8 | 2.6 ± 1.8 | 0.10 |
| 6-month follow-up | 4.0 ± 2.1 ⁎ | 1.9 ± 1.8 † | <0.001 |
Immediately after CRT, the group of patients with favorable outcomes showed significantly larger improvements in the LVEF and SDI than patients without favorable outcome (ΔLVEF 8 ± 6% vs 3 ± 4%, p <0.001; ΔSDI −2.6 ± 2.8% vs −0.3 ± 2.1%, p <0.001; Figure 1 ). Similarly, the improvements in subendocardial and subepicardial LV twist immediately after CRT were more pronounced in patients with favorable outcomes (Δ subendocardial LV twist 1.9 ± 2.8° vs 0.3 ± 2.9°, p = 0.012; and Δ subepicardial LV twist 2.1 ± 2.1° vs −0.2 ± 1.7°, p <0.001; Figure 1 ).
Finally, at 6-month follow-up, further improvements in subendocardial and subepicardial LV twist were observed in patients with favorable outcomes (from 4.0 ± 3.1° to 5.7 ± 2.9°, p = 0.002, for subendocardial LV twist and from 1.9 ± 1.8° to 4.0 ± 2.1°, p <0.001, for subepicardial LV twist; Table 2 ). In Figure 2 , an example of a patient with a favorable outcome (LV reverse remodeling associated with clinical improvement at 6-month follow-up) is shown. In contrast, patients without favorable outcomes showed a trend toward a deterioration of subendocardial LV twist (from 4.5 ± 2.7° to 3.8 ± 2.8°, p = 0.11; Table 2 ) and significant worsening of subepicardial LV twist (from 2.6 ± 1.8° to 1.9 ± 1.8°, p = 0.037; Table 2 ).
