Study Design and Population

In this prospective, observational pilot study conducted at Oslo University Hospital, Rikshospitalet, we recruited 199 patients with AS between May 2005 and April 2009. Patients were assigned to either surgical AVR or optimal medical therapy after evaluation by a multidisciplinary heart team consisting of cardiac surgeons and clinical and interventional cardiologists. After excluding patients with concomitant significant valvular disease ( n = 16), insufficient imaging quality ( n = 13), ventricular paced rhythm ( n = 6), or atrial fibrillation without sufficient rate control ( n = 2), 162 patients were included in the study ( Figure 1 ). All participants gave written informed consent. Medical history including cardiac symptoms and medication was obtained from medical records. New York Heart Association functional class was determined by experienced cardiologists. Coronary artery disease (CAD) was defined as a history of angina pectoris, previous myocardial infarction, or evidence of a stenotic epicardial coronary artery or vein graft on angiography (≥50% diameter stenosis in the absence of collateral perfusion). A standard 12-lead electrocardiogram was obtained at the time of enrollment for automatic measurement of QRS duration.

Flowchart of AS survivors and nonsurvivors.

Data regarding all-cause mortality until August 1, 2012, were obtained from electronic hospital records, synchronized with the Norwegian National Registry. No patients were lost to follow-up. The study complied with the Declaration of Helsinki and was approved by the Regional Committee for Medical Research Ethics.

Echocardiography

All patients underwent a comprehensive transthoracic two-dimensional echocardiographic examination using the Vivid 7 ultrasound system (GE Vingmed Ultrasound, Horten, Norway). Data were analyzed using EchoPAC software (GE Vingmed Ultrasound). AV area (AVA) was calculated using the continuity equation. LVEF was assessed using the Simpson biplane method.

We used valvulo-arterial impedance (Zva), calculated as the sum of systolic arterial pressure and the mean AV transvalvular pressure gradient divided by stroke volume index, as a measure of global afterload. GLS was assessed by speckle-tracking strain and defined as the average of peak negative longitudinal shortening from 16 LV segments using apical four-chamber, two-chamber, and long-axis views ( Figure 2 ). We defined contraction duration as the time from the onset of Q/R on the electrocardiogram to peak negative longitudinal strain in 16 LV segments, and mechanical dispersion was calculated as the SD of contraction durations in the same 16 segments.

Peak strain and mechanical dispersion. Speckle-tracking echocardiography showing longitudinal strain curves in the four-chamber view from patients with AS. White horizontal arrows indicate contraction duration, defined as time from Q/R on the electrocardiogram to peak negative longitudinal strain. Mechanical dispersion was defined as the SD of contraction duration from 16 LV segments. ( Left ) A more homogenous contraction pattern (lower mechanical dispersion) in a survivor, whereas in the AS nonsurvivor ( right ), mechanical dispersion is more pronounced.

Feasibility and Variability Analyses

Only patients with technically adequate echocardiograms for speckle-tracking analysis were included, and 94% of the myocardial segments could be analyzed. Strain analyses were performed by two observers (L.G.K., R.M.A.T.B.) and showed interobserver intraclass correlation coefficients of 0.91 (95% CI, 0.62–0.98; P < .01) and 0.94 (95% CI, 0.75–0.99; P < .01) for mechanical dispersion and GLS, respectively, in 10 randomly selected study patients. Measurements of mechanical dispersion and GLS by the same observer (L.G.K.) 10 months apart in 10 randomly selected study patients showed intraobserver interclass correlation coefficients of 0.83 (95% CI, 0.32–0.96; P < .01) and 0.95 (95% CI, 0.81–0.99; P < .01), respectively.

Statistical Analyses

Continuous data are presented as mean ± SD and were compared using unpaired Student’s t tests or χ ² tests (SPSS version 21.0; SPSS, Chicago, IL). Linear associations were evaluated using multiple regression. Two-sided P values ≤ .05 were considered to indicate statistical significance. Parameters with P values < .20 in univariate comparisons between survivors and nonsurvivors were evaluated using Kaplan-Meier plots and log-rank statistics.

Survival analysis was performed using Cox proportional-hazards regression with AVR as a time-dependent covariate. Variables with P values < .20 for mortality in log-rank tests were assessed using standard univariate Cox regression and in bivariate time-dependent analysis adjusted for AVR. The final multivariate model with AVR as time-dependent covariate was built using both forward and backward manual selection and an automatic stepwise backward procedure, all resulting in the same variable selection. Nested models were compared using likelihood ratio χ ² statistics. We calculated continuous net reclassification improvement and integrated discrimination index between echocardiographic models using the method of Pencina et al. , including the same variables used in the multivariate analysis. The proportional-hazards assumption was tested using Schoenfeld residuals. Survival analyses were performed in R version 3.3.2 (R Foundation for Statistical Computing, Vienna, Austria).

Study Design and Population

In this prospective, observational pilot study conducted at Oslo University Hospital, Rikshospitalet, we recruited 199 patients with AS between May 2005 and April 2009. Patients were assigned to either surgical AVR or optimal medical therapy after evaluation by a multidisciplinary heart team consisting of cardiac surgeons and clinical and interventional cardiologists. After excluding patients with concomitant significant valvular disease ( n = 16), insufficient imaging quality ( n = 13), ventricular paced rhythm ( n = 6), or atrial fibrillation without sufficient rate control ( n = 2), 162 patients were included in the study ( Figure 1 ). All participants gave written informed consent. Medical history including cardiac symptoms and medication was obtained from medical records. New York Heart Association functional class was determined by experienced cardiologists. Coronary artery disease (CAD) was defined as a history of angina pectoris, previous myocardial infarction, or evidence of a stenotic epicardial coronary artery or vein graft on angiography (≥50% diameter stenosis in the absence of collateral perfusion). A standard 12-lead electrocardiogram was obtained at the time of enrollment for automatic measurement of QRS duration.

Flowchart of AS survivors and nonsurvivors.

Data regarding all-cause mortality until August 1, 2012, were obtained from electronic hospital records, synchronized with the Norwegian National Registry. No patients were lost to follow-up. The study complied with the Declaration of Helsinki and was approved by the Regional Committee for Medical Research Ethics.

Echocardiography

All patients underwent a comprehensive transthoracic two-dimensional echocardiographic examination using the Vivid 7 ultrasound system (GE Vingmed Ultrasound, Horten, Norway). Data were analyzed using EchoPAC software (GE Vingmed Ultrasound). AV area (AVA) was calculated using the continuity equation. LVEF was assessed using the Simpson biplane method.

We used valvulo-arterial impedance (Zva), calculated as the sum of systolic arterial pressure and the mean AV transvalvular pressure gradient divided by stroke volume index, as a measure of global afterload. GLS was assessed by speckle-tracking strain and defined as the average of peak negative longitudinal shortening from 16 LV segments using apical four-chamber, two-chamber, and long-axis views ( Figure 2 ). We defined contraction duration as the time from the onset of Q/R on the electrocardiogram to peak negative longitudinal strain in 16 LV segments, and mechanical dispersion was calculated as the SD of contraction durations in the same 16 segments.

Peak strain and mechanical dispersion. Speckle-tracking echocardiography showing longitudinal strain curves in the four-chamber view from patients with AS. White horizontal arrows indicate contraction duration, defined as time from Q/R on the electrocardiogram to peak negative longitudinal strain. Mechanical dispersion was defined as the SD of contraction duration from 16 LV segments. ( Left ) A more homogenous contraction pattern (lower mechanical dispersion) in a survivor, whereas in the AS nonsurvivor ( right ), mechanical dispersion is more pronounced.

Feasibility and Variability Analyses

Only patients with technically adequate echocardiograms for speckle-tracking analysis were included, and 94% of the myocardial segments could be analyzed. Strain analyses were performed by two observers (L.G.K., R.M.A.T.B.) and showed interobserver intraclass correlation coefficients of 0.91 (95% CI, 0.62–0.98; P < .01) and 0.94 (95% CI, 0.75–0.99; P < .01) for mechanical dispersion and GLS, respectively, in 10 randomly selected study patients. Measurements of mechanical dispersion and GLS by the same observer (L.G.K.) 10 months apart in 10 randomly selected study patients showed intraobserver interclass correlation coefficients of 0.83 (95% CI, 0.32–0.96; P < .01) and 0.95 (95% CI, 0.81–0.99; P < .01), respectively.

Statistical Analyses

Continuous data are presented as mean ± SD and were compared using unpaired Student’s t tests or χ ² tests (SPSS version 21.0; SPSS, Chicago, IL). Linear associations were evaluated using multiple regression. Two-sided P values ≤ .05 were considered to indicate statistical significance. Parameters with P values < .20 in univariate comparisons between survivors and nonsurvivors were evaluated using Kaplan-Meier plots and log-rank statistics.

Survival analysis was performed using Cox proportional-hazards regression with AVR as a time-dependent covariate. Variables with P values < .20 for mortality in log-rank tests were assessed using standard univariate Cox regression and in bivariate time-dependent analysis adjusted for AVR. The final multivariate model with AVR as time-dependent covariate was built using both forward and backward manual selection and an automatic stepwise backward procedure, all resulting in the same variable selection. Nested models were compared using likelihood ratio χ ² statistics. We calculated continuous net reclassification improvement and integrated discrimination index between echocardiographic models using the method of Pencina et al. , including the same variables used in the multivariate analysis. The proportional-hazards assumption was tested using Schoenfeld residuals. Survival analyses were performed in R version 3.3.2 (R Foundation for Statistical Computing, Vienna, Austria).

Clinical Characteristics and Echocardiographic Results

Baseline clinical characteristics both stratified according to survival status and for the total cohort are summarized in Table 1 . A total of 162 patients (48% women) with severe AS were included in the study and followed for 37 ± 13 months. Mean age was 75 ± 9 years, and 106 patients (65%) had at least one cardiovascular risk factor (hypertension, diabetes mellitus, or hypercholesterolemia) at inclusion. Most patients ( n = 143 [88%]) were in New York Heart Association class II or III, and 72 (44%) had histories of CAD. Average QRS duration was 103 ± 17 msec. Among the 16 patients with QRS duration > 120 msec, four patients had left bundle branch block. Intraventricular conduction delay was similarly distributed between survivors and nonsurvivors ( P = .23).

All 162 patients were potential candidates for surgery. After evaluation by a multidisciplinary heart team, surgical AVR was performed in 120 patients (74%; 49 with concomitant coronary artery bypass graft), whereas 42 were deemed ineligible for surgery. Reasons for conservative treatment cited in the electronic medical records were sparse symptoms or lack of motivation ( n = 23) and comorbidities ( n = 19, including one death while on the waiting list for AVR). Reasons for ineligibility for surgical AVR were similarly distributed between survivors and nonsurvivors ( P = .20). Five patients received permanent pacemakers during follow-up, all because of atrioventricular conduction disorders. Four of these patients had been treated with AVR. No implantable cardioverter-defibrillator was implanted in the study population.

By echocardiography, all patients had AS, with a mean gradient of 56 ± 23 mm Hg and mean AVA of 0.68 ± 0.22 cm ² ( Table 2 ). Severe AS with normal flow (stroke volume index > 35 mL/m ² ) was present in 69% of patients ( n = 112). Among patients with low-flow AS (stroke volume index ≤ 35 mL/m ² ) 22% ( n = 35) had high gradient (AV mean gradient > 40 mm Hg), while 9% ( n = 15) had the combination of both low flow and low gradient (AV mean gradient ≤ 40 mm Hg). Overall, the patients had LV septal hypertrophy (1.3 ± 0.3 cm), had nondilated left ventricles, and exhibited diastolic dysfunction (mean E/e′ ratio, 18.5 ± 10.8) with mild left atrial enlargement (24 ± 6 cm ² ). Despite normal LVEF (60 ± 11%), LV GLS (−16.2 ± 3.9%) was reduced. Average mechanical dispersion in the total patient population was 64 ± 20 msec ( Figure 3 ). The coefficients of variation for mechanical dispersion and GLS were 0.31 and 0.24, respectively.

Table 2

Echocardiographic findings in 162 patients with AS

Variable	All patients ( N = 162)	Survivors ( n = 125)	Nonsurvivors ( n = 37)	P ∗
AV mean gradient (mm Hg)	56 ± 23	58 ± 23	53 ± 21	.26
AVA (cm 2 )	0.68 ± 0.22	0.69 ± 0.20	0.67 ± 0.26	.62
E/e′ ratio	18.5 ± 10.8	17.6 ± 8.6	21.6 ± 15.5	.09
E/A ratio	1.0 ± 0.8	1.0 ± 0.6	1.1 ± 1.1	.24
Fractional shortening (%)	38 ± 11	40 ± 9	34 ± 13	<.01
GLS (%)	−16.2 ± 3.9	−16.7 ± 3.6	−14.5 ± 4.4	<.01
IVSd (cm)	1.3 ± 0.3	1.3 ± 0.3	1.3 ± 0.2	.96
LA area (cm 2 )	24 ± 6	24 ± 6	25 ± 8	.29
LV CO (L)	5.0 ± 1.1	5.1 ± 1.1	4.6 ± 1.0	.03
LVEF (%)	60 ± 11	62 ± 9	55 ± 15	<.01
LV SV (mL)	75 ± 17	77 ± 17	67 ± 14	<.01
LV SVi (mL/m 2 )	40 ± 9	41 ± 9	36 ± 7	<.01
LVEDD (cm)	5.2 ± 0.7	5.1 ± 0.7	5.5 ± 0.9	.02
Mechanical dispersion (msec)	64 ± 20	61 ± 18	74 ± 24	<.01
TR peak gradient (mm Hg)	32 ± 11	31 ± 10	38 ± 14	<.01
TR velocity (m/sec)	2.8 ± 0.5	2.7 ± 0.4	3.0 ± 0.3	<.01
Zva (mm Hg · m 2 · mL −1 )	5.3 ± 1.2	5.2 ± 1.2	5.7 ± 1.3	.03

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Alterations in Cardiac Deformation, Timing of Contraction and Relaxation, and Early Myocardial Fibrosis Accompany the Apparent Recovery of Acute Stress-Induced (Takotsubo) Cardiomyopathy: An End to the Concept of Transience
2. First-Trimester Fetal Echocardiography: Identification of Cardiac Structures for Screening from 6 to 13 Weeks’ Gestational Age
3. Reduced Myocardial Reserve in Young X-Linked Muscular Dystrophy Mice Diagnosed by Two-Dimensional Strain Analysis Combined with Stress Echocardiography
4. Strain Rate in Children and Young Piglets Mirrors Changes in Contractility and Demonstrates a Force-Frequency Relationship

Stay updated, free articles. Join our Telegram channel

Join