Background
The presence and determinants of left ventricular (LV) dyssynchrony in patients with aortic stenosis (AS) are not clear. The aims of this study were to (1) investigate the presence and determinants of LV dyssynchrony and (2) assess if LV dyssynchrony could improve after aortic valve replacement (AVR) in patients with AS with narrow QRS complexes.
Methods
Twenty healthy subjects and 30 consecutive patients with AS were retrospectively studied. AVR was performed in 19 patients. The time to peak systolic velocity with reference to the QRS complex (Ts), the standard deviation of Ts (Ts-SD), and maximal difference of Ts were measured as the index of LV dyssynchrony in 12 LV segments on Doppler tissue imaging.
Results
Ts-SD (25 ± 17 vs 52 ± 15 msec) and the maximal difference of Ts (70 ± 47 vs 148 ± 38 msec) were significantly greater ( P < .001) in patients with AS than in healthy subjects. Early after AVR (11 ± 4 days), LV dyssynchrony significantly improved with the shortening of Ts-SD (29 ± 14 msec) and the maximal difference of Ts (91 ± 42 msec) ( P < .001). Ts-SD was significantly correlated with estimated LV systolic pressure ( r = 0.53, P < .001) and LV mass index ( r = 0.28, P = .02).
Conclusions
LV dyssynchrony is not uncommon in patients with AS with narrow QRS complexes and is reversible early after AVR, suggesting the favorable effect of afterload reduction on dyssynchronous LV contraction.
Patients with aortic stenosis (AS) are subjected to chronic pressure overload of the left ventricle, leading to hypertrophy, myocardial ischemia, and eventually systolic and diastolic dysfunction. Beneficial effects of aortic valve replacement (AVR) are surgical relief of increased afterload, regression of LV mass, and achievement of a better balance between myocardial oxygen supply and demand. Recently, with the advent of cardiac resynchronization therapy, there is growing concern about the pathophysiologic mechanisms of left ventricular (LV) dyssynchrony regardless of QRS duration. Increased afterload and myocardial ischemia have been demonstrated to induce LV dyssynchrony in animal studies. The association between LV hypertrophy and dyssynchrony was demonstrated in asymptomatic patients with hypertension and normal QRS duration. The pathophysiologic characteristics of AS include increased afterload, myocardial ischemia, and hypertrophy. Consequently, LV dyssynchrony might develop in AS with progression without abnormal electrical conduction. However, the presence of LV dyssynchrony and the effects of AVR on LV dyssynchrony in patients with AS are not well known. Echocardiography with Doppler tissue imaging has emerged as one of the most important noninvasive and readily repeatable tools to assess LV function and dyssynchrony. In addition, Doppler tissue imaging enables the evaluation of LV dyssynchrony with high temporal resolution. Therefore, the aims of this study were to (1) investigate the presence and determinants of LV dyssynchrony and (2) assess if LV dyssynchrony could improve after AVR in patients with AS with narrow QRS complexes and preserved systolic function by Doppler tissue imaging.
Methods
Study Population
We retrospectively studied 30 consecutive patients with isolated moderate to severe AS (peak aortic velocity ≥ 3.0 m/sec) referred to the National Cerebral and Cardiovascular Center and 20 healthy subjects randomly recruited from the community. Patients with renal insufficiency with plasma creatinine levels > 1.5 mg/dL, old myocardial infarctions, angina with significant coronary artery stenosis, more than mild mitral valve disease or aortic regurgitation, pulmonary disease, and atrial fibrillation were excluded from the study. Control subjects had no histories of cardiovascular or systemic diseases but did have normal electrocardiographic and echocardiographic findings. This study was conducted in accordance with institutional ethical guidelines on human research. All subjects gave informed consent before participating in this study.
Echocardiography
Standard and tissue Doppler echocardiographic variables were measured before and 11 ± 4 days after AVR using a 2.5-MHz imaging probe connected to a commercially available cardiac ultrasound system (Vivid 7; GE Vingmed Ultrasound AS, Horten, Norway). LV diastolic and systolic diameters and septal and posterior wall thicknesses were measured using two-dimensional and M-mode echocardiography. On the basis of these measurements, LV fractional shortening was calculated. Transaortic peak velocity was obtained by continuous-wave Doppler echocardiography, and its pressure gradient was calculated using the simplified Bernoulli equation. Estimated LV systolic pressure was obtained as the sum of systolic blood pressure and transaortic pressure gradient. Estimated LV systolic pressure was used as the index of LV afterload. The LV mass index (LVMI) was estimated using the formula of Devereux and Reicheck.
Assessment of Dyssynchrony
For the assessment of LV dyssynchrony, color Doppler tissue imaging was performed to evaluate the long-axis motion of the left ventricle from apical four-chamber, two-chamber, and long-axis views as described by Yu et al. Two-dimensional echocardiography with color-coded Doppler tissue imaging was optimized to obtain an appropriate velocity scale, image sector width and depth, and range of interest. The mean frame rate was >150 frames/sec (6.7-msec interval), which enabled us to analyze myocardial velocity profiles in high temporal resolution. At least three consecutive beats were saved, and the images were analyzed using a customized software package (EchoPAC PC; GE Healthcare, Milwaukee, WI). We placed region of interest at 12 LV segments (six mid and six basal) of the septal, anteroseptal, anterior, lateral, posterior, and inferior walls. Myocardial tissue velocity of each region of interest was reconstructed offline. The time to peak systolic velocity with reference to the QRS complex (Ts) was measured during the ejection phase. Ts was corrected for heart rate using Bazett’s formula and expressed in milliseconds. The ejection phase was determined by the time interval between the onset and the end of LV outflow tract velocity signals derived from pulsed-wave Doppler echocardiography. In each subject, the longest Ts, the shortest Ts, and their difference were measured among 12 LV segments. The standard deviation of Ts (Ts-SD), which has been demonstrated as an index of systolic intraventricular dyssynchrony, was calculated.
Cardiac Catheterization
Coronary angiography was performed in all patients before AVR. The severity of coronary stenosis was quantified in two orthogonal views, and a stenosis was classified as significant if the luminal diameter reduction was >50%.
Statistical Analysis
Statistical analysis was performed using commercially available statistics software (StatView version 5.0; SAS Institute Inc., Cary, NC). Categorical variables are expressed as absolute values and were compared using Fisher’s exact tests between the control and AS groups. Continuous variables are expressed as mean ± SD. Comparisons of mechanical dyssynchrony indices and other parametric echocardiographic measurements among patient groups were performed using one-way repeated analysis of variance with Bonferroni’s correction as appropriate. Univariate linear regression analysis was used to investigate the correlation between two parametric variables. Multivariate linear regression analysis was used to explore the independent determinants of the index of LV dyssynchrony. The results were considered to be significant for P values < .05.
Results
Patient Characteristics
Characteristics of studied subjects are shown in Table 1 . There was no significant difference in age ( P = .45) or gender ( P = .23) distribution between the control and AS groups. The durations of QRS complexes were narrow and did not differ between the control and AS groups. All patients with AS had no significant coronary artery stenosis. AVR was performed in 19 patients, and echocardiographic examinations were performed early after AVR (mean duration after AVR, 11 ± 4 days). All showed reductions in symptoms after AVR. There were no significant changes after AVR in systemic systolic and diastolic blood pressures, but heart rate increased significantly. LV fractional shortening did not change significantly after AVR.
| Variable | Controls ( n = 20) | Patients with AS | |||
|---|---|---|---|---|---|
| All ( n = 30) | No AVR ( n = 11) | Before AVR ( n = 19) | After AVR ( n = 19) | ||
| Age (y) | 66 ± 15 | 69 ± 11 | 71 ± 13 | 68 ± 9 | 68 ± 9 |
| Men/women | 9/11 | 8/22 | 2/9 | 6/13 | 6/13 |
| Body surface area (m 2 ) | 1.51 ± 0.15 | 1.47 ± 0.16 | 1.43 ± 0.19 | 1.48 ± 0.14 | 1.45 ± 0.12 |
| Heart rate (beats/min) | 64 ± 8 | 65 ± 11 | 65 ± 8 | 65 ± 12 | 79 ± 12 ∗,† |
| Systolic blood pressure (mm Hg) | 127 ± 18 | 129 ± 22 | 133 ± 11 | 128 ± 26 | 119 ± 12 |
| Diastolic blood pressure (mm Hg) | 74 ± 11 | 71 ± 13 | 69 ± 9 | 72 ± 15 | 65 ± 7 |
| Symptoms | |||||
| Asymptomatic | 0 | 8 | 3 | 5 | 0 |
| Angina | 0 | 12 | 1 | 11 | 0 |
| Syncope | 0 | 3 | 2 | 1 | 0 |
| Heart failure | 0 | 11 | 4 | 7 | 0 |
| QRS duration (msec) | 83 ± 12 | 90 ± 15 | 91 ± 20 | 90 ± 11 | 86 ± 12 |
| Days after AVR (d) | — | — | — | — | 11 ± 4 |
| LV end-diastolic dimension (mm) | 46 ± 4 | 45 ± 7 | 44 ± 4 | 46 ± 8 | 42 ± 8 |
| LV end-systolic dimension (mm) | 28 ± 4 | 27 ± 8 | 25 ± 3 ∗ | 28 ± 10 | 28 ± 8 |
| Septal wall thickness (mm) | 8.5 ± 1.7 | 12.3 ± 2.0 ∗ | 12.0 ± 2.3 ∗ | 12.4 ± 1.9 ∗ | 11.3 ± 2.0 ∗ |
| Posterior wall thickness (mm) | 9.2 ± 1.5 | 12.4 ± 2.0 ∗ | 12.4 ± 2.8 ∗ | 12.5 ± 1.5 ∗ | 11.2 ± 1.7 ∗,† |
| LV fractional shortening (%) | 38 ± 7 | 40 ± 9 | 44 ± 5 ∗ | 39 ± 10 | 35 ± 9 |
| LVMI (g/m 2 ) | 113 ± 31 | 180 ± 60 ∗ | 184 ± 58 ∗ | 178 ± 63 ∗ | 146 ± 45 † |
| LV end-systolic volume (mL) | 33 ± 9 | 45 ± 20 | 37 ± 11 | 49 ± 22 | 39 ± 23 |
| LV end-diastolic volume (mL) | 76 ± 16 | 97 ± 28 ∗ | 91 ± 22 | 100 ± 31 | 90 ± 36 |
| Biplane LVEF (%) | 56 ± 4 | 57 ± 9 | 60 ± 7 | 55 ± 9 | 57 ± 7 |
| Transaortic peak PG (mm Hg) | — | 96 ± 31 | 76 ± 19 | 106 ± 31 | — |
| Transaortic mean PG (mm Hg) | — | 50 ± 25 | 36 ± 9 | 61 ± 29 | — |
| Aortic valve area (cm 2 ) | — | 0.62 ± 0.18 | 0.67 ± 0.21 | 0.59 ± 0.16 | — |
LV Dyssynchrony before and after AVR
The indexes of LV dyssynchrony were measured before and after AVR. Average Ts, Ts-SD, and the maximal difference of Ts were all significantly greater in patients with AS than in control subjects, suggesting the presence of LV dyssynchrony in patients with AS ( Table 2 ). Before AVR, Ts was shorter in the anteroseptal, anterior, and septal segments and longer in the posterior, lateral, and inferior segments. After AVR, the difference in Ts among 12 LV segments was decreased. An example of regional myocardial tissue velocity curves obtained by color Doppler tissue imaging is shown in Figure 1 . Figure 2 shows Ts in all 12 segments in the control subjects and patients with AS. There was marked prolongation of Ts at the 10 LV segments (basal and mid anteroseptal, lateral, posterior, inferior, and septal segments) before AVR, which normalized after AVR compared with the corresponding values obtained from control subjects. Early after AVR (mean, 11 ± 4 days), the degree of LV dyssynchrony significantly decreased with significant shortening of average Ts, Ts-SD, and maximal difference of Ts ( Table 2 , Figure 3 ).
| Variable | Controls ( n = 20) | Patients with AS | |||
|---|---|---|---|---|---|
| All ( n = 30) | No AVR ( n = 11) | Before AVR ( n = 19) | After AVR ( n = 19) | ||
| Longest Ts | 183 ± 55 | 274 ± 44 ∗ | 260 ± 47 ∗ | 283 ± 40 ∗ | 218 ± 51 † |
| Shortest Ts | 114 ± 26 | 126 ± 26 | 124 ± 11 | 127 ± 19 | 127 ± 28 |
| Average Ts | 144 ± 37 | 198 ± 35 ∗ | 199 ± 47 ∗ | 198 ± 26 ∗ | 163 ± 31 † |
| Maximum difference of Ts | 70 ± 47 | 148 ± 38 ∗ | 135 ± 35 ∗ | 155 ± 39 ∗ | 91 ± 42 † |
| Ts-SD | 25 ± 17 | 52 ± 15 ∗ | 48 ± 13 ∗ | 54 ± 15 ∗ | 29 ± 14 † |
