Aortic stenosis leads to left ventricular hypertrophy and accumulation of fibrillar collagens. The analysis of integrated backscatter (IBS) parameters provides information on ultrasonic myocardial properties.
The study population consisted of 58 patients with aortic stenosis. They were followed up for an average 18 ± 5 months after aortic valve replacement (AVR). Traditional transthoracic echocardiography and analysis of IBS reflectivity were performed before AVR and during the control visit after AVR.
A significant reduction in left ventricular mass index, a significant increase in the mean cyclic variation of IBS, and a decrease in absolute end-diastolic IBS intensity were observed after AVR.
These data suggest improvements in ultrasonic myocardial properties after AVR. Preoperative analysis of IBS parameters might provide additional information for predicting left ventricular reverse remodeling in patients a mean of 1.5 years after AVR for aortic stenosis.
Aortic stenosis (AS) is a common valvular heart disease associated with life-threatening complications and a high mortality rate in symptomatic patients in its natural history. Pressure overload in patients with AS leads to left ventricular (LV) hypertrophy associated with deposits of new sarcomeres and the accumulation of fibrillar collagens throughout the free wall and interventricular septum of the heart. Myocardial fibrosis has been shown to increase myocardial stiffness and to promote abnormalities of diastolic and systolic cardiac function and intramyocardial perfusion. Although aortic valve replacement (AVR) dramatically improves the clinical courses of patients with AS by relieving the high-pressure aortic gradient, many factors affecting LV remodeling after AVR are still unknown.
Ultrasonic myocardial tissue characterization by integrated backscatter (IBS) has been successfully used for the differentiation of various myopathies from normal myocardium. The analysis of IBS indices provides information on myocardial fibrosis and ultrasonic myocardial properties. A number of experimental and clinical studies have shown that IBS characterization of myocardial texture correlates with tissue collagen and water content and with the degree of myocardial hypertrophy in patients with hypertension, but there is no strong evidence of the usefulness of IBS in patients with AS.
The aim of this study was to assess the diagnostic and prognostic value of different parameters of IBS measurement in relation to LV reverse remodeling and systolic and diastolic LV function changes following AVR in patients with AS.
Seventy-three consecutive symptomatic patients with isolated AS and clinical and hemodynamic indications for AVR surgery according to American College of Cardiology, American Heart Association, and European Society of Cardiology guidelines were enrolled in the study. Exclusion criteria were mitral valve disease, more than mild aortic regurgitation, medical history of myocardial infarction, chronic atrial fibrillation, and malignant or accelerated arterial hypertension. Fifteen patients did not undergo the post-AVR follow-up examination, because of new atrial fibrillation, pacemaker rhythm (rhythm other than sinus rhythm), and prosthesis dysfunction. They were excluded from further analysis. The final study population consisted of 58 patients (40 men, 18 women; mean age, 68 ± 10 years). The mean follow-up period after AVR surgery was 18 ± 5 months. AVR was performed in the Cardiosurgery Department of the Medical University of Gdansk. Thirty-three biologic and 25 mechanical prostheses were implanted (mean label size, 23 ± 2 mm).
Traditional transthoracic echocardiography and measurement of IBS parameters were performed before AVR. The follow-up examination was performed during the control visit with the same echocardiographic settings as in the initial study, on average 1.5 years after AVR. The study protocol was approved by the ethics committee of the Medical University of Gdansk, and written informed consent was obtained from all patients.
All patients underwent two-dimensional and Doppler echocardiography. Recordings and measurements were obtained according to the recommendations of the American Society of Echocardiography. Measurements of LV end-diastolic diameter, interventricular septal thickness, and posterior wall thickness were made at the beginning of the QRS complex. LV mass (LVM) was calculated using Devereux’s formula and normalized to body surface area as the LVM index (LVMI). LV hypertrophy was considered to be present when LVMI was >116 g/m 2 in men and >104 g/m 2 in women. Relative wall thickness (RWT) was defined as the ratio (2 × posterior wall thickness)/LV end-diastolic diameter, and values ≤ 0.43 were considered normal. Patterns of LV remodeling in patients with AS were classified according to LVMI and RWT values as normal LVM and geometry (LVMI < 116 g/m 2 in men, LVMI < 104 g/m 2 in women, and RWT < 0.43), concentric remodeling (LVMI < 116 g/m 2 in men, LVMI < 104 g/m 2 in women, and RWT > 0.43), concentric hypertrophy (LVMI > 116 g/m 2 in men, LVMI > 104 g/m 2 in women, and RWT > 0.43), and eccentric hypertrophy (LVMI > 116 g/m 2 in men, LVMI > 104 g/m 2 in women, and RWT < 0.43). For the purpose of this study, we defined reverse LV remodeling as the change of preoperative type of LV remodeling by at least one pattern toward normal LVM and geometry.
Systolic LV Function
We assessed LV systolic function by evaluation of ejection fraction (EF), fractional shortening (FS), midwall FS (mwFS), and circumferential end-systolic wall stress (cESS). EF was measured using the biplane Simpson’s method, and mwFS was calculated using a previously validated formula by de Simone et al. Circumferential end-systolic wall stress was estimated at the LV midwall level using a cylindrical model described by Gaasch et al. adjusted to the intraventricular LV pressure at the end of systole, which was calculated as a sum of the cuff systolic blood pressure and maximal Doppler-driven transvalvular gradient through stenotic native valve or valve prosthesis after surgery.
Diastolic LV Function
Doppler parameters of mitral inflow such as the mitral E and A waves (early and late diastolic filling velocities), mitral E-wave deceleration time, and the isovolumic relaxation time, were measured using pulse Doppler. Pulmonary venous flow measurements included peak systolic velocity, peak diastolic velocity, and peak atrial reversal velocity. According to the Canadian consensus recommendations for the measurement and reporting of diastolic dysfunction by echocardiography, the following patterns of LV filling were classified: normal pattern of LV filling, delayed relaxation, pseudonormal filling, and restrictive filling. The average values of early mitral annular diastolic velocity (e′) were obtained from both the septal and lateral sites and averaged for further evaluation. Subsequently, the ratio of mitral peak velocity of early filling to early diastolic mitral annular velocity (E/e′) was calculated. Improvement of diastolic function was defined as the change of preoperative pattern of LV diastolic function by at least one pattern toward normal LV filling parameters.
AS Severity Assessment
Peak and mean transaortic valve pressure gradients (PGs) and velocity-time integral (VTI) of transaortic velocity were measured using continuous-wave Doppler from different windows. The highest velocity was used for tracing the VTI. Effective aortic valve area (AVA) was calculated using the continuity equation and then indexed to body surface area. The ratio of LV outflow tract (LVOT) velocity to transaortic velocity was also calculated (VTI LVOT /VTI Ao ). Aortic valve prostheses were evaluated on the same basis as native valves. Patient-prosthesis mismatch (PPM) was diagnosed when the effective orifice area (EOA) was <0.9 cm 2 /m 2 .
Ultrasonic Tissue Characterization
To assess IBS signal, two regions of interest were chosen in the parasternal long-axis view: the mid septum and the mid posterior wall. The mean value of IBS was acquired from the region of interest for each imaged frame during the cardiac cycle and averaged for end-diastole and end-systole for all cardiac cycles throughout the cine loop. The absolute intensity of IBS at end-diastole (IBS ed ) and at end-systole (IBS es ) was obtained for both the septum and the posterior wall. The cyclic variation of IBS (CVIBS) was calculated as the difference between end-diastolic and end-systolic values (CVIBS = IBS ed − IBS es ) for the septum and the posterior wall ( Figure 1 ). Additionally, we calculated the index of CVIBS (CVIBS index ) at the septum and at the posterior wall, which were calculated by use of the formula [(IBS ed − IBS es )/IBS ed ] × 100. A detailed IBS acquisition methodology was described previously. All examinations were recorded on magneto-optical disks and analyzed offline by experienced echocardiographers (M.F., A.K., and R.G.) using a Philips Sonos 5500 revision D.2 with a 3s transducer (Philips Medical Systems, Andover, MA) by means of a commercially available acoustic densitometry software package. According to a previous study, a mean value of CVIBS > 6 dB was considered normal. The mean IBS parameters (mean CVIBS, mean IBS ed , and mean CVIBS index ) were defined as average values obtained from both the septum and the posterior wall. Additionally, we evaluated the sensitivity and specificity of preoperative mean IBS ed , mean CVIBS, and mean CVIBS index different thresholds for LV reverse remodeling occurrence after aortic valve surgery. Differences of preoperative and postoperative values were also calculated for all parameters (Δ).
Reproducibility Data of IBS Evaluation
Fifteen randomly selected IBS evaluations were made twice by two echocardiographers to assess interobserver and intraobserver variability. Agreement between two measurements was considered as ±0.1 dB. Intraobserver and interobserver variability were good (κ = 0.74 and κ = 0.67, respectively).
For the statistical analysis, Statistica for Windows version 9.0 (StatSoft, Tulsa, OK) was used. Continuous data are expressed as mean ± SD. Paired t tests were used to compare preoperative and postoperative parameters. One-way analysis of variance followed by Bonferroni’s post hoc test was used to assess the statistical difference between more than two groups of continuous parameters. To measure the strength of the relation between CVIBS and other echocardiographic parameters, Pearson’s correlation coefficient was calculated. The relation between clinical improvement after AVR and echocardiographic parameters was evaluated using multiple logistic regression analysis. Receiver operating characteristic (ROC) curve analysis was generated by MedCalc version 11.2.1 ( www.medcalc.com ) to test the predictive discrimination of IBS parameters. A p value < .05 was considered significant. The κ statistic was used to assess intraobserver and interobserver variability for IBS measurements.
Preoperative and postoperative clinical data of the patients enrolled in the study are outlined in Table 1 .
|Variable||Before AVR||After AVR||P|
|Heart rate (beats/min)||73 ± 12||71 ± 10||NS|
|Systolic blood pressure (mm Hg)||130 ± 14||141 ± 18||<.05|
|Diastolic blood pressure (mm Hg)||76 ± 11||80 ± 12||<.05|
|Mean blood pressure (mm Hg)||94 ± 12||101 ± 13||<.05|
|Weight (kg)||75 ± 11||78 ± 11||<.001|
|Body mass index (kg/m 2 )||26.6 ± 2.9||27.9 ± 3.3||<.001|
Echocardiographic preoperative and postoperative data obtained during the initial and follow-up examinations are outlined in Table 2 . Forty-three patients (74%) had LV hypertrophy (mean LVM, 262 ± 88 g; mean LVMI, 143 ± 45 g/m 2 ). There was no correlation between LV hypertrophy and echocardiographic parameters of AS severity such as AVA, maximal PG, or VTI LVOT /VTI Ao . The prevalence of different LV remodeling patterns was as follows: normal geometry in three patients (5%), concentric remodeling in 20 patients (35%), concentric hypertrophy in 28 patients (48%), and eccentric hypertrophy in seven patients (12%). There were no significant differences of AS severity parameters among particular patterns of LV remodeling.
|Variable||Before AVR||After AVR||P|
|IVSd (mm)||14 ± 3||13 ± 2||<.001|
|LVDd (mm)||47 ± 9||44 ± 6||<.001|
|PWd (mm)||13 ± 2||12 ± 2||<.001|
|LVDs (mm)||31 ± 10||27 ± 6||<.001|
|RWT||0.59 ± 0.17||0.55 ± 014||NS|
|LVEF (%)||72 ± 13||77 ± 9||<.05|
|FS (%)||36 ± 9||40 ± 8||<.05|
|mwFS (%)||12.2 ± 3.1||14.1 ± 3.1||<.001|
|cESS (kdyne/cm 2 )||207 ± 52||165 ± 32||.001|
|E (cm/s)||77 ± 23||85 ± 24||<.05|
|A (cm/s)||86 ± 34||89 ± 27||NS|
|E/A||1.2 ± 1.0||1.1 ± 0.6||NS|
|DT (ms)||250 ± 95||254 ± 78||NS|
|S (cm/s)||53 ± 16||62 ± 15||<.001|
|D (cm/s)||41 ± 16||45 ± 12||NS|
|S/D||1.5 ± 0.6||1.5 ± 0.5||NS|
|Ar (cm/s)||29 ± 6||28 ± 6||NS|
|e′ (cm/s)||7.4 ± 2.4||9.6 ± 2.0||<.05|
|E/e′||11.6 ± 6.1||9.3 ± 3.5||<.001|
|Maximal PG (mm Hg)||87 ± 13||34 ± 22||<.001|
|Mean PG (mm Hg)||56 ± 22||19 ± 14||<.001|
|AVA (cm 2 )||0.7 ± 0.3||1.6 ± 0.6||<.001|
|AVA index (cm 2 /m 2 )||0.4 ± 0.1||0.8 ± 0.3||<.001|
|VTI LVOT /VTI Ao||0.24 ± 0.1||0.48 ± 0.1||<.001|
|LVM (g)||262 ± 88||198 ± 58||<.001|
|LVMI (g/m 2 )||143 ± 45||106 ± 29||<.001|
The prevalence of different patterns of LV diastolic dysfunction was as follows: normal diastolic function in 11 patients (20%), delayed relaxation in 36 patients (62%), pseudonormal filling in six patients (10%), and restrictive filling in five patients (8%). There was no significant difference of AS severity parameters in particular patterns of diastolic dysfunction.
The preoperative mean CVIBS was 5.8 ± 1.7 dB, mean IBS ed was 36.2 ± 5.0 dB, and mean CVIBS index was 17.1 ± 5.0%. Significantly lower values of mean CVIBS before surgery were observed in the subgroup of patients with eccentric LV hypertrophy in comparison with patients with LV concentric remodeling or hypertrophy. Additionally, significantly higher mean IBS ed was observed in patients with eccentric LV hypertrophy in comparison with patients with normal LVM and dimensions ( Table 3 ). There were no significant differences of mean IBS values in patients with different diastolic filling patterns. However, in the subgroup of patients with mean CVIBS ≤ 6 dB before AVR in comparison with subjects with mean CVIBS > 6 dB, the initial maximal velocity of e′ was significantly lower (6.9 ± 2.5 vs 8.3 ± 2.0 cm/s, P < .05), and the E/e′ ratio was significantly higher (13.0 ± 7.2 vs 9.6 ± 2.9, P < .05). Both CVIBS and CVIBS index at the septum and posterior wall showed positive significant correlations with EF, FS, and e′ and negative significant correlations with cESS and LVMI. There were no significant correlations between all evaluated IBS indices and AVA or mean or maximal PG ( Table 4 ).
|Mean IBS ed (dB)||32.4 ± 3.3||36.2 ± 4.8||36.8 ± 5.2||41.2 ± 4.7 ∗|
|Mean CVIBS (dB)||6.4 ± 1.6||6.3 ± 1.7||5.7 ± 1.7||3.9 ± 1.8 †|
|Mean CVIBS index (%)||17.2 ± 2.9||18.7 ± 5.4||16.9 ± 4.7||12.9 ± 4.9|
|Septal CVIBS||0.32 ∗||0,26 ∗||0.22||−0.36 ∗||0.04||−0.34 ∗||0.03||0.02||0.50 ∗|
|Posterior wall CVIBS||0.35 ∗||0.28 ∗||0.17||−0.40 ∗||0.12||−0.27 ∗||0.10||0.06||0.42 ∗|
|Mean CVIBS||0.36 ∗||0.29 ∗||0.20||−0.39 ∗||0.04||−0.34 ∗||0.08||0.05||0.48 ∗|
|Septal IBS ed||0.22||0.20||0.21||−0.14||−0.05||−0.01||0.06||0.00||0.19|
|Posterior wall IBS ed||0.29 ∗||0.26||0.22||−0.24||0.00||−0.09||0.05||0.05||0.29 ∗|
|Mean IBS ed||0.27 ∗||0.24||0.22||−0.21||−0.01||−0.12||−0.01||0.02||0.25|
|Septal CVIBS index||0.27 ∗||0.20||0.17||−0.26 ∗||0.03||−0.35 ∗||−0.07||0.02||0.49 ∗|
|Posterior wall CVIBS index||0.36 ∗||0.29 ∗||0.16||−0.44 ∗||0.17||−0.37 ∗||−0.11||0.10||0.48 ∗|
|Mean CVIBS index||0.29 ∗||0.23||0.17||−0.32 ∗||0.07||−0.36 ∗||−0.08||0.08||0.50 ∗|