|Home | About | Journals | Submit | Contact Us | Français|
We hypothesized that LV segmental dyssynchrony, quantified by paradoxical systolic wall thinning, determines changes in global LV performance in a model of canine right ventricular pacing-induced dyssynchrony and the response to CRT.
Quantification of left ventricular (LV) dyssynchrony is important to assess the impact of cardiac resynchronization therapy (CRT).
Seven pentobarbital-anesthetized open-chest dogs had LV pressure-volume relations and mid-LV short-axis echocardiographic speckle tracking radial strain imaging during right atrial pacing (RA), right ventricular pacing (RV) to simulate left bundle branch block, and CRT using RV plus either LV free wall (CRTfw) and apical (CRTa) pacing. The area under the segmental LV time-radial strain positive and negative curves defined global thickening and thinning, respectively. Dyssynchrony was defined as the maximum time difference between earliest and latest peak segmental positive strain among 6 radial sites.
RA had minimal dyssynchrony (58±40 ms). RV induced both dyssynchrony (213±67 ms, *p<0.05) and reduced LV stroke work (SW)(67±51 mJ* p<0.05). CRTfw and CRTa decreased dyssynchrony (116±47 and 50±34 ms, respectively, * vs. RV) but only CRTa restored LV SW to RA levels. RV decreased global thickening (129±87%•msec) compared to RA (258±133%• ms p<0.05), whereas CRTfw and CRTa restored regional thickening to RA levels (194±83 and 230±76% •ms, respectively). The sum of thickening and thinning during RV (230±88 vs. 258±133%•ms*) correlated (r=0.98) with RA thickening, suggesting that all the loss of LV function was due to thinning.
Dyssynchrony causes proportional changes in regional LV wall thinning and global LV stroke work that were reversed by CRT, suggesting that dyssynchrony impairs LV systolic function by causing paradoxical regional wall thinning and that CRT effectiveness can be monitored by its reversal. Thus, monitoring paradoxical regional thinning reversal may be used to define CRT effectiveness.
Quantification of left ventricular (LV) dyssynchrony by echocardiography is important for the assessment of baseline cardiac performance and to quantify cardiac resynchronization therapy (CRT) effectiveness (1–14). However, the quantitative contribution that regional dyssynchrony and its reversal during CRT make toward global LV contraction effectiveness is unclear.
Tissue Doppler strain analysis can be applied for evaluation of ischemia, LV and right ventricular (RV) function, and mechanical dyssynchrony (1, 3, 6, 8, 12, 13, 15). However, visualization of tissue Doppler phase shift analysis requires a minimal angle of incidence to measure tissue velocity. Thus, these studies are usually restricted to assessment of longitudinal LV movement, greatly limiting its ability to assess global LV motion. Speckle tracking of B-mode gray-scale 2D echocardiographic images does not require specific angles of incidence of the myocardium relative to the echo sensor. These speckle-tracking techniques have been validated against sonomicrometry in vivo and magnetic resonance imaging in vitro (16–18). Recent data using magnetic resonance imaging have shown that circumferential strain is more sensitive to assess cardiac dyssynchrony than longitudinal motion (19, 20). Furthermore, we documented that regional radial strain analysis can quantify ventricular dyssynchrony using both tissue Doppler radial strain (21–23) and speckle tracking radial strain in humans (24). In this study, we hypothesized that LV segmental dyssynchrony, quantified by paradoxical systolic wall thinning, determines changes in global LV performance during dyssynchrony and the response to CRT. To assess LV systolic wall thickening and thinning we used echocardiographic speckle tracking techniques to quantify regional strain and its impact on global LV performance in our intact acute canine model (21, 23).
Seven mongrel dogs, weighing 21.0±1.5 kg were studied after an overnight fast. The protocol was approved by the institutional animal care and use committee and conformed to the AHA position on research animal use. All dogs were anesthetized with sodium pentobarbital (30mg/kg induction; 1.0mg/kg/h with intermittent boluses, as needed) and mechanically ventilated. A 6F 11-pole multi-electrode conductance catheter (Webster Laboratories, Irvine, California) and a LV micromanometer catheter (MPC-500, Millar, Houston, Texas) were placed for LV pressure-volume analysis, as previously described (21). After a median sternotomy, a snare occluder was placed around the inferior vena cava to transiently alter preload. The pericardium was opened and epicardial pacing wires (A & E Medical Corp. Farmingdale, NJ, USA) were placed on the right atrium (RA), RV free-wall near the anterior infundibulum, LV mid-free-wall near the mid-posterior-lateral wall, and LV apex for multi-site stimulation. The pericardium was re-opposed and positive end-expiratory pressure (PEEP) applied to re-expand the lungs. Afterward, 5 cm H2O PEEP was applied for the remainder of the experiment. Fluid resuscitation was performed prior to starting the protocol to restore apneic LV end-diastolic volume to pre–sternotomy values.
LV pressure, volume and electrocardiogram signals were digitized at 150 Hz and stored on disk for offline analysis. LV peak systolic pressure, stroke volume (SV) and stroke work (SW) were used to assess global LV performance. Stroke work was calculated as the integral of the pressure-volume loop.
An echocardiographic system (Aplio SSA-770A, Toshiba Medical Systems Corp, Tokyo, Japan) was used to obtain images with a 3.0 MHz transducer directly applied to the heart. Digital B-mode gray-scale 2D and tissue Doppler cine loops from 3 consecutives beats were obtained at end-expiratory apnea from mid-LV short axis view at depths of 8 cm using a fixed transducer position. Mid-LV short-axis views were selected with the papillary muscle as a consistent internal anatomic landmark, and great care was taken to orient the image to the most circular geometry possible. The temporal resolution of the tissue Doppler system was frame rates of 49 Hz with a pulse repetition frequency of 4.5 kHz. The precise spatial resolution of the speckle tracking algorithm was estimated to be in the range of 0.5 to 1 mm. Gain settings were adjusted for echo imaging to optimize endocardial definition. Offline analysis of radial strain was then performed on digitally stored images (ApliQ, Toshiba Corp.).
The speckle tracking analysis was used to generate regional LV strain (25, 26) from echocardiographic images. Strain-time waveforms were generated for frame-by-frame movement of stable patterns of natural acoustic markers present in ultrasound tissue images over the cardiac cycle as previously described by our group (24). Briefly, a circular region of interest was traced on the endocardial and epicardial border of the mid-LV short axis view. Speckles within the region of interest were tracked in subsequent frames by the imaging software. The location shift of these speckles from frame to frame, representing tissue movement, provided the spatial and temporal data. Myocardial thickening/thinning was calculated as change in length/initial length between specific endocardial and epicardial speckles (ΔL/Lo). Myocardial thickening was represented as positive strain, color-coded yellow, and thinning was represented as negative strain, color-coded blue, and superimposed on conventional 2D images. The software divides the short-axis image into 6 equal segments. Thickening and thinning values from multiple circumferential points were calculated and data averaged into 6 segmental time-strain curves, as previously validated in humans (24) and dogs (27).
Thickening and thinning were quantified by speckle tracking at the same 6 regions of interest during RA pacing as heart rate control, RV, RA-RV-LV free wall pacing (CRTfw) and RA-RV-LV apex pacing (CRTa), as previously described in this model (23). Time-to-peak strain from each of 6 time-strain curves was determined with dyssynchrony defined as the difference between earliest and latest segments (21). Global radial strain was calculated as averaged radial strain from 6 segments. Peak global strain was used as a marker of global contractility. Global thickening was calculated as the area under the curve of the global positive strain. RA served as baseline control, while RV served as maximal dyssynchrony. We calculated the thinning as the sum of areas under the curves of negative individual segment strains (Figure 1b). Global performance was quantified as LV stroke work. We assumed that if the observed reduction in global performance was explained by regional dyssynchrony, then thickening would be associated with LV stroke work and the sum of thickening plus thinning during RV should be correlated to thickening during RA.
All measurements were made during apnea with 5 cm H2O PEEP. RA pacing was performed at frequencies 5–10 min−1 above the intrinsic rhythm. RA was defined as normal ventricular contraction. All succeeding ventricular pacing studies were done with sequential pacing at an A–V delay of 30 ms, preventing atrial fusion beats but also eliminated atrial augmentation of LV filling. High RV free-wall pacing was used to induce an LBBB-like contraction pattern. We then compared the impact of CRTfw and CRTa on regional and global LV performance. The order of CRTfw and CRTa was alternated among animals. Pacing was maintained for > 30 seconds before measurements were made. In practice, hemodynamic stability usually took < 15 seconds to occur. Between each ventricular-paced rhythm, animals were returned to RA pacing and all hemodynamic variables allowed to stabilized before the next step was initiated. Three heart cycles were analyzed per pacing mode.
Data are expressed as mean ± SD. Repeated measures analysis of variance was used for comparisons among different pacing modalities. A paired sample Student's t test was used for the statistical comparison of parametric values. These paired t-tests are made without correction for multiple comparisons. Since RV, CRTa and CRTfw did not have atrial contribution to diastolic filling, LV end-diastolic volume was lower than in RA. Thus, global LV performance comparisons were made only between RV, CRTa and CRTfw. Significance was determined as p < 0.05. Inter-observer variability was assessed in ten randomly selected studies for segmental time-to-peak strain and global peak strain and was calculated as the ratio (%) of the difference between the values obtained by each observer (expressed as absolute value) divided by the mean of the two values. Intra-observer variability was calculated by a similar approach.
The maximum time difference from earliest to latest peak strain among 6 segments was minimal with RA (58 ± 40 ms). RV displayed early septal thickening and opposing free wall thinning, typically seen in human LBBB (25). Maximal time strain difference increased during RV (213 ± 67ms, p<0.05 vs. RA) (Table 1, Figure 3), while both CRTa and CRTfw reduced maximal time difference (50 ± 34 and 116 ± 47ms, CRTa and CRTfw, respectively, p<0.05 vs. RV) (Figure 3). RV resulted in less LV wall thickening (258±133 %•ms vs. 129±87 %•ms, p<0.05), while both CRTa and CRTfw restored thickening to RA levels (230±76 %•ms and 194±83 %•ms, CRTa and CRTfw respectively, p<0.05 vs. RV). Systolic thinning during RV occurred for 101±41 %•ms. The sum of thickening and thinning during RV correlated with thickening during RA (230±88 %•ms and 258±133 %•ms, respectively p<0.05, r= 0.98), suggesting that all the LV function loss was due to thinning.
Global radial strain, defined as the average among 6 radial segments decreased during RV compared to RA (22 ± 9% vs. 13 ± 8%, p<0.05) and was restored during both CRTa and CRTfw (22 ± 6% and 19 ±7 %, respectively, p<0.05 vs. RV).
Intra- and inter-observer variability for segmental time-to-peak strain expressed as the mean percent error (absolute difference/mean) was 4 ± 8% and 4 ± 8%, respectively (n=56 segments), and for global peak strain was 13 ± 9% and 15 ± 9%, respectively (n=10 studies).
RV pacing-induced dyssynchrony caused decrements in global LV function and regional contraction, as quantified by global hemodynamics and regional strain. Both CRTa and CRTfw reduced LV dyssynchrony but only CRTa improved both regional and global LV function relative to RV. Thus, regional contraction abnormalities, as quantified by paradoxical systolic wall thinning, parallel impaired global LV performance during pacing-induced dyssynchrony but only partially explain the effects of CRT from different resynchronization sites on global LV performance. Importantly, regional dyssynchrony as assessed by segmental thickening relative to thinning of a mid-axis cross-sectional image varies in proportion with global LV function during RA and RV. Because our model allowed us to compare normal contraction with dyssynchronous contraction, a luxury not available under clinical conditions of CRT, we documented that the combined thickening and thinning of RV equaled the normal thickening time of RA, suggesting that dyssynchrony-associated global LV impairment can be quantified by the relative degree of systolic thinning. CRTa and CRTfw resulted in similar improvements in thickening compared to RV but different quantitative effects of global LV performance. Thus, factors other than mid-plane radial contraction such as torsion and longitudinal strain analysis may be needed to determine the subsequent improvement in global LV performance during CRT.
These observations have clinical relevance. Radial thickening plays a major role in LV ejection. Myocardial contraction not only shortens cardiac myocytes, it thickens them. Our data suggests that the changes in global LV performance can be quantified by dyssynchronous strain activity because contraction dyssynchrony across the entire myocardium can be used as quantitative markers of dyssynchrony and LV global performance. Speckle tracking assesses strain in a non-directional fashion, unlike tissue Doppler imagining, allowing one to analyze isolated regional thinning and the timing of peak thickening during systole across the entire myocardium. Although MRI analysis of contraction dyssynchrony is powerful in assessing the impact of regional contraction dyssynchrony on global performance (28), its data acquisition is cumbersome and its post-acquisition data analysis laborious.
Our data agree with previous studies showing that dyssynchronous contraction increases radial strain differences across LV segments when assessed by M-mode, tissue Doppler imaging and MRI scanning (11, 19, 22, 28). We recently demonstrated that speckle tracking radial strain can quantify dyssynchrony and predict immediate and long term responses to CRT in humans (24).
Our findings that LV global and regional functions are differentially affected by pacing site agree with previous investigators (21, 23, 29). LV function was better if the CRT pacing was from the LV apex. These data agree with our previous study (21) and those of others (19). Presumably, apical pacing activates the intact His-Purkinje system before it travels from the LV apex upwards toward the base. We previously showed that patients in whom LV lead position was concordant with the site of latest mechanical activation by speckle tracking radial strain had a greater increase in ejection fraction from baseline than patients with discordant lead position (24). Furthermore, a time-to-peak strain difference in humans between septal to free wall peak cutoff value of >130ms identified those candidates who would improve LV function following CRT (24). All our animals except one had a pacing-induced dyssynchrony > 130ms and improved their performance with CRT. In the one who did not, neither CRTa nor CRTfw improved synchrony. These data support the hypothesis that CRT can only improve performance if baseline dyssynchrony is above some minimal threshold.
There are two major limitations of this study. First, we studied an intact canine model without intrinsic conduction defects or impaired contractility. We (23), and others (29), have documented that LV apical CRT is superior to LV free wall CRT in terms of global LV performance and resynchronization in a model of pacing-induced dyssynchrony. However, similar apical pacing superiority has not been reported in human CRT studies. Dyssynchrony usually occurs through structural myocardial defects, which themselves are not similarly distributed across all CRT candidates. Our data does demonstrate, however, that different pacing site CRT may have differing effects on global LV performance. Second, speckle-tracking echocardiography is dependent on frame rates, as well as image resolution. Low acquisition frame rates degrade assessment of regional myocardial motion and its subsequent strain rate analysis. In contrast, increasing the frame rate reduces scan-line density, which reduces image resolution (25, 26). Suffoletto et al. (24) found that frame rates in the range of 30 to 90 Hz with a mean of 65 Hz suitable for speckle-tracking analysis. Accordingly, in our study we used a mean frame rate of 49 Hz suitable for speckle tracking analysis.
The authors hypothesized that left ventricular (LV) segmental dyssynchrony, quantified by paradoxical systolic wall thinning, determines changes in global LV performance during canine right ventricular pacing-induced dyssynchrony and the response to cardiac resynchronization therapy (CRT). Dyssynchrony caused proportional changes in regional LV wall thinning and global LV stroke work that were reversed by CRT, suggesting that dyssynchrony impairs LV systolic function by causing paradoxical regional wall thinning and that CRT effectiveness can be monitored by its reversal. Thus, monitoring paradoxical regional thinning reversal may be used to define CRT effectiveness. This proposed parameter will need to be tested against exciting echocardiographic parameters and tested clinically.
The authors thank Lisa Gordon and Don Severyn for the expert technical assistance.
This study was supported in part by NIH awards HL04503, HL067181 and HL073198.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.