Background
Impaired left ventricular (LV) myocardial deformation is associated with adverse outcome in patients with severe aortic stenosis (AS). The aim of this retrospective study was to assess the impact of transcatheter aortic valve implantation (TAVI) on the recovery of myocardial mechanics and the influence of postprocedural aortic regurgitation (AR).
Methods
Speckle-tracking echocardiography was used to assess multidirectional myocardial deformation (longitudinal and circumferential strain) and rotational mechanics (apical rotation and twist) before and at midterm follow-up after TAVI. Predictors of myocardial recovery, defined as a ≥20% relative increase in the magnitude of global longitudinal strain compared with baseline, were examined.
Results
Sixty-four patients (median age, 83 years; interquartile range, 77–86 years) with severe AS and high surgical risk (mean European System for Cardiac Operative Risk Evaluation score, 20 ± 13%) were evaluated. Overall, LV longitudinal deformation was impaired at baseline compared with controls. At 5 ± 3 months after TAVI, LV longitudinal deformation had significantly improved only in the group of patients with baseline LV ejection fractions (LVEF) ≤ 55%: global longitudinal strain from −9.7 ± 3.7% to −11.8 ± 3.2% ( P = .05), longitudinal strain rate from −0.44 ± 0.14 sec −1 to −0.57 ± 0.16 sec −1 ( P = .001), and early diastolic strain rate from 0.38 ± 0.17 sec −1 to 0.49 ± 0.18 sec −1 ( P = .01). In patients with normal LVEFs, LV twist was supraphysiologic at baseline and normalized after TAVI (from 16.1 ± 6.9° to 11.9 ± 6.2°, P = .004). In patients with baseline LVEFs ≤ 55%, circumferential deformation was impaired before TAVI and improved after TAVI. Baseline LVEF (odds ratio, 0.56 per 10% increment; P = .02) and global longitudinal strain (odds ratio, 0.65 per absolute 1% increment; P < .001) were significant predictors of myocardial recovery. LV mass, volumes, and longitudinal strain failed to favorably remodel in patients with post-TAVI important AR (defined as new mild post-TAVI AR or moderate or severe post-TAVI AR [either preexisting or new AR]).
Conclusions
TAVI restores LV function toward more physiologic myocardial mechanics in both normal- and depressed-LVEF groups. Patients with lower systolic function derive the most benefit in terms of longitudinal reverse remodeling. Postprocedural AR adversely affects LV structural and functional remodeling.
Aortic stenosis (AS) causes chronic pressure overload on the left ventricle, leading to concentric hypertrophy, subendocardial ischemia, myocardial fibrosis, impaired diastolic filling, and potentially systolic dysfunction. Transcatheter aortic valve implantation (TAVI) is a novel therapy for patients with severe AS at high risk for open-heart surgery. Patients who have undergone TAVI are older and have more serious comorbidities than patients who have been referred for conventional surgical aortic valve replacement (AVR). One potential disadvantage of TAVI is an increased incidence of postprocedural aortic regurgitation (AR), which is an independent predictor of short- and long-term mortality and which may have a negative impact on LV myocardial recovery.
Speckle-tracking echocardiography (STE) allows the quantitative angle-independent assessment of myocardial mechanics in different axes and rotations. STE provides strain and strain rate (SR) measurements that are more sensitive markers of subtle alterations in global and regional myocardial function. In the setting of AS, studies have shown progressive impairment in global longitudinal systolic strain (GLS) and SR, proportional to the severity of AS, despite preserved left ventricular (LV) ejection fraction (LVEF). Reduction in GLS predicts a worsened prognosis in this population.
Therefore, the assessment of multidirectional myocardial recovery by STE after TAVI in this elderly population with a high burden of comorbidities warrants further characterization. We sought to determine the intermediate-term impact of TAVI on myocardial mechanics by using comprehensive quantification of LV longitudinal, circumferential, and rotational deformation before and after TAVI in a population of symptomatic patients with severe AS. Our secondary objective was to assess the influence of postprocedural AR on the recovery of myocardial mechanics.
Methods
Study Design
Study Population
This retrospective study consisted of patients undergoing TAVI for symptomatic severe AS at a single center (Toronto General Hospital, University of Toronto). Patients were included in this study if transthoracic echocardiograms obtained before TAVI and at medium-term follow-up (between 2 and 12 months) were available for review. Exclusion criteria for this study were (1) poor endocardial tracking with speckle-tracking echocardiographic analysis in at least two adjacent myocardial segments and (2) the presence of atrial fibrillation during the echocardiographic study. Twenty-one healthy patients ≥60 years of age studied previously using the same echocardiographic protocol served as the control group. The study protocol was approved by the local institutional research ethics board.
TAVI
Eligibility criteria for TAVI were the presence of symptomatic severe native or prosthetic valve stenosis with an aortic valve area ≤ 1.0 cm 2 and/or mean systolic aortic gradient > 40 mm Hg. In the setting of LV systolic dysfunction and low-flow, low-gradient AS, the severity of AS was confirmed by low-dose dobutamine stress echocardiography. All potential candidates for TAVI were considered to have an excessively high risk for death with conventional open AVR, with an estimated operative mortality risk of >15%, as determined by a multidisciplinary team of experts.
Clinical Data
Demographic characteristics, comorbidities, previous cardiac procedures, logistic European System for Cardiac Operative Risk Evaluation score, and functional status were prospectively collected in a database.
Echocardiography
Transthoracic echocardiographic and Doppler studies were done before and after TAVI at our institution. Two-dimensional echocardiographic, Doppler, and Doppler tissue imaging parameters were measured in accordance with published guidelines from the American Society of Echocardiography. LV volumes and LVEF were estimated using Simpson’s biplane method. Aortic valve area was calculated using the continuity equation. Peak and mean systolic transaortic gradients were calculated using the simplified Bernoulli equation.
AR
Before patients underwent TAVI, native aortic valve regurgitation was ascertained by integrating information from the following available semiquantitative parameters: AR jet width of the LV outflow tract, AR jet cross-sectional area in the LV outflow tract, AR jet density, pressure half-time of AR, and diastolic flow reversal in the descending aorta. The categorization of pre-TAVI AR into mild, moderate, and severe groups was made according to recommendations of the American Society of Echocardiography. After the TAVI procedure, color, pulsed-wave, and continuous-wave Doppler imaging was performed to semiquantitatively evaluate the severity of post-TAVI AR. The categorization of post-TAVI paravalvular AR was based mainly on the proportion of the circumference of the aortic prosthetic ring occupied by the regurgitant jet in the parasternal short-axis view. Trivial paravalvular AR was defined as a pinpoint jet. We adopted the following classification scheme for the circumferential extent of the AR jet, which has been proposed by the Valve Academic Research Consortium–2 consensus document: (1) mild, <10%; (2) moderate, 10% to 29%; and (3) severe, ≥30%. Pulsed-wave signals of diastolic flow reversal in the descending thoracic aorta was also used to differentiate post-TAVI AR, with the following criteria: (1) mild, absent or brief early diastolic flow reversal; (2) moderate, intermediate findings between mild and severe AR; and (3) severe, prominent and holodiastolic flow reversal. Finally, the density of the AR jet on continuous-wave Doppler was used to help differentiate the cases of post-TAVI AR: (1) mild, incomplete or faint AR jet density, or (2) moderate or severe, dense AR jet density. In cases of both valvular and paravalvular regurgitation seen in the postprocedural echocardiographic studies, the grading of AR reflected the summation of total regurgitation.
In analyzing post-TAVI valvular regurgitation, we hypothesized that any new post-TAVI AR potentially affects LV recovery, because this valvular regurgitation represents a new insult, mimicking the hemodynamic changes of acute AR. New mild post-TAVI AR contrasts with preexisting mild AR (i.e., a patient with both pre-TAVI mild AR of the original aortic valve and post-TAVI mild AR of the new aortic bioprosthesis), in which the LV cavity presumably does not need to make any major hemodynamic adaptation to an unchanged degree of AR. Accordingly, we dichotomized patients into two subgroups: unimportant post-TAVI AR and important post-TAVI AR. We defined unimportant post-TAVI AR as (1) no or trivial post-TAVI AR or (2) preexisting mild post-TAVI AR (i.e., at least mild pre-TAVI AR and ongoing post-TAVI mild AR). Important post-TAVI AR was defined as (1) new mild post-TAVI AR (i.e., no or trivial pre-TAVI AR that becomes mild AR after TAVI) or (2) moderate or severe post-TAVI AR (either preexisting or new AR). All images were digitally stored for offline analyses.
Myocardial Mechanics
Multidirectional myocardial mechanics were compared in patients with normal and abnormal LVEFs, with a cutoff LVEF of >55% used to define the normal systolic function group. Quantitative assessment of LV subendocardial mechanics by STE was performed before and after TAVI using Velocity Vector Imaging version 3 (Siemens Medical Solutions USA, Mountain View, CA). Apical four-, three-, and two-chamber views were used to obtain longitudinal strain and SR, as previously described. Parasternal short-axis planes were used to obtain circumferential strain, SR, rotational angles, and maximal instantaneous basal to apical angle difference (net LV twist). Physiologic counterclockwise apical rotation was expressed as a positive angle. Global peak systolic strain and SR and early diastolic SR values were derived from the time-strain and SR curves, averaging the 16 myocardial segments.
Intraobserver and Interobserver Variability
Offline two-dimensional strain evaluations were done by a single operator (F.P.) blinded to patients’ clinical information. For intraobserver variability, 10 randomly selected studies were reanalyzed once by the same observer several months after the initial analysis. A second experienced observer, blinded to previously obtained data, analyzed the same patients and the exact same loops for the assessment of interobserver variability.
Statistical Analyses
Categorical variables are expressed as frequencies and percentages. Continuous variables are summarized as mean ± SD or median (interquartile range), depending on the normality of distribution. Echocardiographic parameters before and after TAVI were compared using McNemar’s test for categorical variables and the paired t test or Wilcoxon’s signed-rank test for continuous variables, as appropriate. Univariate logistic regression models with relevant clinical and echocardiographic variables were used to identify predictors of favorable functional myocardial recovery after TAVI, defined as ≥20% relative increase in the magnitude of GLS compared with baseline (e.g., increase from −15% [baseline] to at least −18% [after TAVI]).
Comparisons of the change in echocardiographic parameters after TAVI between patients with unimportant AR and those with important AR after TAVI were performed using analysis of covariance with the absolute difference from baseline as the outcome and preprocedural value and LVEF as the covariates. Mean percentage change from baseline is also reported to facilitate comparison between groups in clinically relevant terms. Intraobserver and interobserver variability was assessed by the intraclass correlation coefficient and the coefficient of variation. A level of significance of .05 was set for all analyses. All statistical analyses were conducted using SPSS version 20 (SPSS, Inc., Chicago, IL).
Results
Study Population
Among the 121 patients who underwent TAVI for severe AS from 2007 to 2012, and who had available preprocedural and mid-term follow-up transthoracic echocardiograms, 64 were eligible for this study. The excluded individuals consisted of 28 patients because of the presence of atrial fibrillation at the time of echocardiography and 29 patients because of poor endocardial tracking with the Velocity Vector Imaging software (caused mainly by insufficient endocardial definition or tracking issues during the cardiac cycle). Baseline characteristics are depicted in Table 1 . The median age was 83 years (interquartile range, 77–86 years) and 58% were men. The mean logistic European System for Cardiac Operative Risk Evaluation risk estimate was 20 ± 13%. There was no significant difference between the 64 study patients and the 57 excluded patients. The comparison of patients in the normal- and abnormal-LVEF groups showed no significant differences in age (median, 82 years [interquartile range, 77–86 years] vs 83 years [interquartile range, 77–87 years]; P = .96), prevalence of hypertension (76% vs 74%, P = .88), and diabetes (22% vs 30%, P = .45). However, patients in the abnormal-LVEF group had more severe functional limitation (New York Heart Association class III or IV in 100% vs 80%, P = .04), higher logistic European System for Cardiac Operative Risk Evaluation score (25 ± 15% vs 16 ± 12%, P = .02), and a trend toward a greater prevalence of coronary artery disease (74% vs 49%, P = .05), compared with those patients with preserved LVEFs, respectively.
| Clinical characteristic | All patients ( n = 64) | LVEF > 55% ( n = 41) | LVEF ≤ 55% ( n = 23) | P ∗ |
|---|---|---|---|---|
| Age (y) | 83 (77–86) | 82 (77–86) | 83 (77–87) | .96 |
| Men | 37 (58%) | 20 (49%) | 17 (74%) | .05 |
| BMI (kg/m 2 ) | 22 ± 5 | 22 ± 6 | 23 ± 4 | .55 |
| NYHA class III or IV | 56 (87%) | 33 (80%) | 23 (100%) | .04 |
| Arterial hypertension | 48 (75%) | 31 (76%) | 17 (74%) | .88 |
| Dyslipidemia | 43 (67%) | 31 (76%) | 12 (52%) | .06 |
| Diabetes mellitus | 16 (25%) | 9 (22%) | 7 (30%) | .45 |
| Coronary artery disease | 37 (58%) | 20 (49%) | 17 (74%) | .05 |
| Peripheral vascular disease | 8 (13%) | 5 (12%) | 3 (13%) | 1.00 |
| Cerebral vascular disease | 6 (9%) | 2 (5%) | 4 (17%) | .18 |
| Previous AVR | 3 (5%) | 1 (2%) | 2 (9%) | .29 |
| Mitral regurgitation (moderate or greater) | 8 (13%) | 5 (12%) | 3 (13%) | 1.00 |
| Chronic lung disease | 6 (9%) | 5 (12%) | 1 (4%) | .29 |
| Renal insufficiency (GFR ≤ 50 mL/min) | 22 (34%) | 18 (44%) | 4 (17%) | .03 |
| Logistic EuroSCORE (%) | 20 ± 13 | 16 ± 12 | 25 ± 15 | .02 |
TAVI
Transapical and transfemoral approaches were used in 39 (61%) and 23 (36%) patients, respectively. The direct transaortic approach was required in two patients. Edwards SAPIEN valves were implanted in 52 patients (81%) and Medtronic CoreValve devices in 12 patients (19%).
Echocardiography
Follow-up transthoracic echocardiographic studies were performed 5.6 ± 3.5 months after the procedures, with no significant difference between the preserved- and depressed-LVEF subgroups (5.7 ± 3.6 months vs 5.6 ± 3.4 months, P = .90; Table 2 ). The mean aortic valve area increased (from 0.67 ± 0.15 to 1.69 ± 0.41 cm 2 [ P < .001] in the preserved-LVEF subgroup and from 0.70 ± 0.22 to 1.71 ± 0.38 cm 2 [ P < .001] in the depressed-LVEF subgroup), accompanied by a significant decrease in the mean transaortic pressure gradient (from 52 ± 16 to 12 ± 5 mm Hg [ P < .001] in the preserved-LVEF subgroup and from 45 ± 16 to 11 ± 5 mm Hg [ P < .001] in the depressed-LVEF subgroup) after TAVI. In addition, significant reduction in LV systolic volume (from 74 ± 33 to 59 ± 36 mL, P = .01), LV mass index regression (from 138 ± 37 to 120 ± 33 g/m 2 , P = .008), and improvement in LVEF (from 45 ± 9% to 51 ± 12%, P = .04) were observed in the abnormal-LVEF group. The proportion of patients with at least moderate mitral regurgitation at baseline and at follow-up was 13% and 11%, respectively.
