Experimental And Clinical Study On Evaluation Of Left Ventricular Diastolic Function By Using Spot Tracking Echocardiography

Objective:Left ventricular (LV) diastolic dysfunction has emerged as one of the primary factors that determines long-term survival in patients with coronary artery disease (CAD), providing prognostic information. Conventional two-dimensional and Tissue Doppler echocardiography have been widely proposed for noninvasive assessment of LV diastolic function in CAD patients. Because of the known limitations of TDI, however, mostly associated with its angle dependence, it has been gradually replaced by the recently developed two-and three-dimensional speckle-tracking echocardiography (2D-STE and 3D-STE), which intrinsically circumvent most of the drawbacks of Doppler echocardiography. The purpose of the present study was to investigate the diagnostic value of 2D-STE and 3D-STE for estimating left ventricular diastolic function in CAD patients with preserved left ventricular ejection fraction (LVEF).Methods:Eighty-four CAD patients with normal LVEF and thirty age- and sex-matched healthy controls were prospectively recruited. All participants underwent conventional echocardiography,2D-STE and 3D-STE. The three standard parastemal LV apical (apical 4-chamber, apical 2-chamber, and apical 3-chamber) views and the parastemal short-axis views at the level of the mitral valve, mid-papillary, and apex were acquired. Then longitudinal, radial and circumferential strain rate during the period of LV early diastolic filling (L-SRe, R-SRe and C-SRe) and strain rate during the period of LV late diastolic filling (L-SRa, R-SRa and C-SRa) were analyzed using 2D-STE. Global strain values of 3D-STE, including global longitudinal strain (GLS), global circumferential strain (GCS), global area strain (GAS) and global radial strain (GRS), were analyzed offline using a commercially available software (4D Auto LVQ) in the apical four chamber full-volume image. Left ventricular end-diastolic pressure (LVEDP) was invasively obtained by left heart catheterization. Pearson’s correlation coefficient was used for estimating the correlation between echocardiographic parameters and LVEDP. The area under the receiver operating characteristic curve (ROC) was calculated to determine the capability of the above echocardiographic parameters for detecting left ventricular diastolic dysfunction.Results:According to LVEDP, eighty-four CAD patients were further divided into two groups:group I (LVEDP ≤ 15mmHg, n=37); group Ⅱ (LVEDP> 15 mmHg, n=47). In speckle-tracking echocardiographic analyses, compared with the controls, CAD patients showed lower levels in the global early diastolic strain rate (SRe) in the longitudinal, circumferential, and radial directions of myocardial deformation, and displayed higher ratios of E/SRe in three deformations. All four 3D-STE indices (GLS、GCS、GAS and GRS) were significantly lower in CAD patients than in the controls. Within the two CAD groups, patients with elevated LVEDP showed significantly lower levels in SRe and higher levels in E/SRe of the above three deformations; and these 3D-STE variables were significantly impaired in patients with elevated LVEDP. Pearson’s correlation analysis revealed that LVEDP levels correlated positively with E/E’ sep (r=0.60, P<0.01), R-SRe (r=0.38, P<0.01), E/L-SRe (r=0.64, P<0.01), E/C-SRe (r=0.57, P<0.01), GLS (r=0.60, P<0.01), GCS (r=0.51, P<0.01) and GAS (r=0.55, P<0.01). It correlated negatively with L-SRe(r=-0.53, P<0.01), E/R-SRe(r=-0.38, P<0.01), C-SRe(r=-0.45, P<0.01), and GRS (r=-0.58, P<0.01). ROC analysis revealed that series of speckle-tracking echocardiographic variables could be used to predict LVEDP>15 mmHg in CAD patients (areas under the curve:E/L-SRe 0.74; E/C-SRe 0.74; C-SRe 0.70; L-SRe 0.69; E/R-SRe 0.68; R-SRe 0.65; GLS 0.78; GRS 0.77; GCS 0.75; GAS 0.74). However, the areas under the curve for these above speckle-tracking echocardiographic variables were not significantly larger when compared with the common E/E’sep ratio (area under the curve:0.84, P>0.05).Conclusion:Speckle-tracking echocardiographic variables of 2D-STE and 3D-STE are closely related with LVEDP and can be used to predict LV diastolic dysfunction in CAD patients with preserved LVEF. However, these speckle-tracking echocardiographic indices may not be superior to the commonly used E/E’ sep ratio in CAD patients with preserved LVEF.Objective:Global strain rate during early diastole (SRe) measurements from two-dimensional speckle-tracking echocardiography (2D-STE) provide direct information on intrinsic myocardial segments and play an important role in assessing the left ventricular (LV) diastolic dysfunction. The aim of this study was to investigate the capability of 2D-STE in assessing LV early diastolic dysfunction (LVEDD) in pacing-induced canine heart failure.Methods:Pacing systems were implanted in 13 adult mongrel dogs, and continuous rapid right ventricular pacing (RVP,240 beats/min) was maintained for 2 weeks, resulting in early left ventricular diastolic dysfunction model. Echocardiography and left heart catheterization were performed at baseline, during pacing, and at 2 weeks after RVP initiation. Changes in heart morphology were observed by light microscopy and scanning electron at 2 weeks after RVP onset. All dogs underwent conventional echocardiographic examination, including the left atrial diameter, end-diastolic interventricular septal thickness, relative wall thickness, LV internal diameter at end-diastolic (LVIDd), end-diastolic LV postwall thickness, LV ejection fraction (LVEF) and fractional shortening (LVFS), the isovolumetric relaxation time (IVRT), and the E/E’ratio et al. The obtained strain rates from the left ventricular included global early diastolic strain rate (SRe) and global late diastolic strain rate (SRa) in the longitudinal (L-SRe, L-SRa), circumferential (C-SRe, C-SRa), and radial directions (R-SRe, R-SRa). Then we calculated the ratios of peak early diastolic mitral inflow measurement (E) to SRe in the longitudinal, circumferential, and radial directions (E/L-SRe, E/C-SRe and E/R-SRe, respectively). The following variables were measured from cardiac catheterization: systolic blood pressure, diastolic blood pressure, LV end-diastolic pressure (LVEDP), LV end-systolic pressure, and the peak rates of LV pressure increase (LV +dp/dtmax) and decrease (LV -dp/dtmax). Conventional echocardiographic,2D-STE, and LV pressure parameters were measured, and correlations were analyzed during the progressive development of diastolic dysfunction produced by RVP. Receiver operating characteristic (ROC) curves of strain rate imaging parameters were analyzed to identify predictors of LV diastolic dysfunction.Results:In 8 mongrel dogs that completed the study, we found that the onset of LVEDD occurred 3 days after RVP initiation. Compared with the baseline, LVEDD was detected by a significant increase in LVEDP (7.9 ±3.8 mmHg vs.16.3±5.7 mmHg, P<0.05) at the third day after RVP onset, and a significant decrease in LV ±dp/dtmax. RVP progressively decreased LV ± dp/dtmax and increased LVEDP. The end point of our study was evaluated by a significant increase in LVEDP (7.9 ± 3.8 vs.28.1 ± 7.3 mmHg, P<0.01) during the second week of RVP. Light and transmission electron microscopy showed myocardial vacuolar degeneration and mitochondrial swelling in the left ventricular at 2 weeks after RVP onset. Most of the two-dimensional speckle-tracking echocardiographic parameters changed relative to the baseline values after 1 or 3 days of pacing. The L-SRe significantly decreased after the first day of RVP (1.8 ±0.6 S-1 vs.1.2 ± 0.3 S-1, P<0.01) and steadily decreased through the remainder of the pacing period. The C-SRe and R-SRe fell significantly after the third day of RVP. These ratios of E/L-SRe, E/C-SRe and E/R-SRe steadily increased at the third day of RVP. There were no significant differences on the SRa in the longitudinal, circumferential and radial directions. Pearson’s correlation analysis revealed that parameters of conventional echocardiography and 2D-STE showed moderate correlation with LV pressure parameters, including E/E’ sep (r=0.58, P<0.01), IVRT(r=0.55, P<0.01), L-SRe (r=-0.58, P<0.01), E/L-SRe (r=0.65, P<0.01), and R-SRe (r=0.53, P<0.01) et al. Meanwhile, these above indices also showed moderate correlation with LV -dp/dtmax. ROC curves analysis showed that these indices of conventional echocardiography and strain rate imaging could effectively predict LV diastolic dysfunction (area under the curve:E/E’sep 0.78; IVRT 0.66; L-SRe 0.84; E/L-SRe 0.80; R-SRe 0.80).Conclusion:Our data highlight the notion that 2D-STE is a sensitive and accurate technique that can be used for assessing LV early diastolic dysfunction in pacing-induced canine heart failure. The novel indices, including L-SRe, E/L-SRe and R-SRe et al, show close correlation with LV pressure parameters, and can serve as effective, noninvasive LVEDD markers.