A relevant proportion of patients, classified as severe aortic stenosis on the basis of valve area ≤1 cm 2 , have a mean transvalvular gradient ≤40 mm Hg, despite a preserved left ventricular ejection fraction (LGSAS). We assessed the clinical and hemodynamic impact of transcatheter aortic valve implantation in patients with symptomatic LGSAS at high risk for surgery or inoperable, according to the type of percutaneous valve implanted. Ninety-five patients received an Edwards SAPIEN valve (Edwards Lifesciences, Irvine, California) and 51 received a Medtronic CoreValve (Medtronic, Inc., Minneapolis, Minnesota). The hemodynamic performance of the 2 valves was similar in term of final transvalvular gradients (10 mm Hg, p = 0.069). Early mortality rate was 7% and was not different between the 2 valves (p = 0.73). During follow-up, cardiovascular mortality rate was similar between groups, and valve type was not a predictor of outcome (p = 0.72). Estimated survival by Kaplan–Meier at 2 years was 70%. At multivariate analysis, life-threatening or major bleeding, postprocedural aortic insufficiency, and acute kidney injury were the major predictors of an adverse outcome. In patients with LGSAS treated by transcatheter aortic valve implantation, the use of balloon-expandable versus self-expandable valves resulted in similar hemodynamic, early, and long-term clinical outcomes.
Severe aortic stenosis (SAS) is defined by current guidelines as an aortic valve area (AVA) ≤1 cm 2 and a mean gradient (MG) >40 mm Hg in the presence of a normal cardiac output. However, a relevant proportion of patients, classified as SAS on the basis of AVA ≤1 cm 2 , have a low MG despite a preserved left ventricular ejection fraction (LVEF). This hemodynamic subset has recently been recognized as a distinct clinical entity and termed low-gradient severe aortic stenosis (LGSAS). Symptomatic patients with LGSAS who are treated conservatively show higher mortality rates compared to those who undergo surgical aortic valve replacement. Even if transcatheter aortic valve implantation (TAVI) is an alternative treatment for high-risk or inoperable patients with symptomatic SAS, only few data exist on whether patients presenting with symptomatic LGSAS may benefit from TAVI. In this subset of patients, TAVI may be an attractive alternative to surgical aortic valve replacement because of the reduced invasiveness and the superior hemodynamic profile of transcatheter heart valve prostheses. We aimed to determine the clinical and hemodynamic impact of TAVI in patients with symptomatic LGSAS according to the type of percutaneous valve implanted, mainly Medtronic CoreValve Revalving System (Medtronic, Inc., Minneapolis, Minnesota) and the Edwards SAPIEN valve (Edwards Lifesciences, Irvine, California).
Methods
This is a retrospective analysis of a prospective multicentre database that includes all patients with symptomatic SAS who underwent TAVI from June 2007 to February 2013 (n = 1,010) involving 5 tertiary centers. All patients were deemed inoperable or at high surgical risk for conventional surgery by a multidisciplinary team consisting of interventional cardiologists and cardiothoracic surgeons. Included in this study were all consecutive patients with symptomatic SAS (AVA ≤1.0 cm 2 ), low MG (MG ≤40 mm Hg), and preserved LVEF (LVEF ≥50%). After selection, 146 patients fulfilled inclusion criteria of this study. The study cohort complies with the Declaration of Helsinki and was approved by the local Ethics Committees, and all patients provided informed written consent for the procedures.
All 146 patients with LGSAS underwent a comprehensive medical history and physical examination. Dyspnea was evaluated by New York Heart Association (NYHA) class. Demographic data, including age, gender, body mass index, and body surface area, and comprehensive clinical data, including history of hypertension, diabetes mellitus, dyslipidemia, coronary artery disease, peripheral arterial disease, previous myocardial infarction, and coronary artery bypass surgery, were collected. Society of Thoracic Surgery scores were calculated with an online risk calculator ( http://riskcalc.sts.org/STSWebRiskCalc273/ ).
Comprehensive baseline transthoracic echocardiograms included left ventricle end-diastolic volume, LVEF calculated with the Simpson method, transvalvular pressure gradient determined by the Bernoulli formula, and AVA calculated by the continuity equation. Postprocedural parameters of left ventricle and prosthesis function were assessed by Doppler echocardiography 24 to72 hours after TAVI and then checked at follow-up.
TAVI was performed using standard techniques as previously described. Vascular access was transfemoral or trans-subclavian using the Medtronic CoreValve Revalving System and transfemoral, transapical, or transaortic using the Edwards Sapien valve. Aortic regurgitation was assessed using aortography immediately after TAVI.
Adverse events were assessed inhospital, and regular clinical follow-up was performed at 1, 6, and 12 months and then at regular 12-month intervals by means of a clinical visit and echocardiographic evaluation. All suspected events were adjudicated by an unblinded clinical event committee comprising a cardiac surgeon and a interventional cardiologist. Baseline clinical and procedural characteristics and all follow-up data were entered into a dedicated database, held at an academic clinical trials unit (Division of Cardiology, University of Turin, Italy) responsible for central data audits and maintenance of the database.
Clinical end points were defined according to the criteria proposed by the Valve Academic Research Consortium 2. Primary end point was all-cause mortality and cardiovascular death at 30 days. Secondary end points included cerebrovascular events (stroke and transient ischemic attack), myocardial infarction, bleeding, acute renal failure, and vascular complications at 30 days. In addition, we assessed all-cause mortality and cardiovascular death at the last follow-up. NYHA functional class status was assessed at baseline and the last follow-up.
Quantitative data are presented as means ± standard deviations and categorical variables as counts and percentages of patients in each category. Normality of quantitative variables was tested with the Shapiro–Wilk test. Quantitative variables were compared between the 2 groups with independent-samples Student t test in case of normality, with the Wilcoxon rank-sum test otherwise. Categorical variables were compared between the 2 groups with the chi-square or Fisher’s exact test, as appropriate. Survival curves were estimated with the Kaplan–Meier method, and the log-rank test was used for comparison between groups. The effect of clinical and echocardiographic variables on survival at 30 days was assessed with univariable logistic regression, and clinically relevant variables with a p value ≤0.1 were included in the multivariable logistic regression model using the backward method to identify independent predictors of 30-day mortality. Cox proportional hazards regression models were used to assess the cumulative all-cause and cardiovascular mortality in long-term follow-up. Clinically relevant variables with a p value ≤0.1 on univariable analysis were included in the multivariable Cox regression model using the backward method to identify independent predictors of mortality in patients with LGSAS.
Using a greedy, nearest neighbor 1:1 matching algorithm on the logit of the propensity score (using calipers of width equal to 0.2 times the standard deviation of the logit of the propensity score), study groups were further matched for baseline data of aortic valve area, moderate or relevant aortic insufficiency, and STS score, obtaining 35 patients treated by CoreValve and 35 patients treated by Edwards valve. Survival curves for cumulative all-cause and cardiovascular mortality in long-term follow-up were estimated with the Kaplan–Meier method, and the Klein and Moeschberger test for matched samples was used for comparison between groups. Two-sided p values <0.05 were considered statistically significant. All statistical analyses were completed with SPSS (version 19.0, SPSS Inc, Chicago, Illinois) and SAS 9.2 (SAS Institute Inc., Cary, North Carolina) for Windows.
Results
Of 1,010 patients, 146 (14%) had LGSAS. Baseline characteristics and 2-dimensional echo data of the study population are presented in Table 1 . Procedural characteristics are presented in Table 2 . The median out-of-hospital clinical follow-up was 13 months (interquartile range 7 to 26). Early results are presented in Table 3 . During follow-up, cardiovascular mortality was not different among groups and valve type was not a predictor of different outcome (p = 0.72). Figure 1 shows survival curves for total and cardiovascular mortality. The 2-year actuarial mortality was 30%.
