Author + information
- Received February 6, 2011
- Revision received May 2, 2011
- Accepted May 6, 2011
- Published online July 1, 2011.
- Marco Mussardo, MD⁎,
- Azeem Latib, MD⁎,†,
- Alaide Chieffo, MD⁎,
- Cosmo Godino, MD⁎,†,
- Alfonso Ielasi, MD⁎,
- Micaela Cioni, MD‡,
- Kensuke Takagi, MD⁎,†,
- Giedrius Davidavicius, MD⁎,
- Matteo Montorfano, MD⁎,
- Francesco Maisano, MD‡,
- Mauro Carlino, MD⁎,
- Annalisa Franco, MD∥,
- Remo D. Covello, MD∥,
- Pietro Spagnolo, MD§,
- Antonio Grimaldi, MD‡,
- Ottavio Alfieri, MD‡ and
- Antonio Colombo, MD⁎,†,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Antonio Colombo, Interventional Cardiology Unit, San Raffaele Scientific Institute, EMO-GVM Centro Cuore Columbus, Via Buonarroti 48, 20145 Milan, Italy
Objectives The aim of this study was to analyze the short-term outcomes after transcatheter aortic valve implantation with the Edwards Sapien THV (ESV), compared with the Sapien XT THV (SXT) (Edwards Lifesciences, Irvine, California).
Background The SXT has been recently commercialized in Europe, but there are no studies analyzing the efficacy and safety of SXT, compared with ESV.
Methods All consecutive patients (n = 120) who underwent transcatheter aortic valve implantation in our center via the transfemoral approach with either ESV (n = 66) or SXT (n = 54). Valve Academic Research Consortium endpoints were used.
Results Mean age was 80 ± 8 years, and mean Logistic-European System for Cardiac Operative Risk Evaluation was 24.9 ± 17.0. The ilio-femoral artery minimal lumen diameter was smaller in patients treated with the SXT (7.27 ± 1.09 mm vs. 7.94 ± 1.08 mm, p = 0.002). Device success was high in both groups (96.3% vs. 92.4%, p = 0.45). Major vascular events were 3-fold lower in the SXT group (11.1% vs. 33.3%, relative risk: 0.40, 95% confidence interval: 0.28 to 0.57; p = 0.004). Life-threatening and major bleeding events were not significantly different between groups (18.5% vs. 27.3% and 35.2% vs. 40.9%, respectively). The SXT group had a lower 30-day Valve Academic Research Consortium combined safety endpoint (20.4% vs. 45.5%; relative risk: 0.44, 95% confidence interval: 0.24 to 0.80; p = 0.004). The 30-day mortality was 1.7% (n = 2). At 30 days, mean transaortic gradient was approximately 10 mm Hg in both groups and the aortic regurgitation was mild-to-moderate in 70.2% of SXT and 76.3% of ESV.
Conclusions The new SXT valve has the same short-term performance as the ESV but seems to be associated with a lower risk of major vascular complications and thus has a broader clinical application.
Transcatheter aortic valve implantation (TAVI) is superior to conventional medical therapy for inoperable symptomatic patients with severe aortic stenosis (1). The improvement of technology and the introduction of newer delivery systems that require smaller sheath sizes have made the procedure safer and applicable to a wider population of patients. The large size of the sheaths required for valve delivery is the most important limitation of this technique. Recently, the new Sapien XT (Sapien XT THV [SXT], Edwards Lifesciences, Inc., Irvine, California) and NovaFlex delivery system received CE-mark and became commercially available in Europe. This new delivery system uses smaller-sized catheters than the previous RetroFlex delivery system.
In the published data there are many studies that demonstrate the efficacy and safety of the Edwards Sapien THV (ESV) (Edwards Lifesciences), but there are no studies that compare it with the new SXT valve (1–5). Thus, we analyzed our experience with transfemoral TAVI with the 2 different Edwards Sapien valves.
We included all consecutive patients undergoing TAVI in our center via the transfemoral approach with the ESV from November 2007 to April 2010 and SXT implanted since the device became commercially available in April 2010 until December 2010. We compared patients treated with ESV and SXT and analyzed periprocedural and short-term outcomes. Clinical follow-up was performed by office visit at 30 days after the procedure.
We used the recently described endpoint definitions of the Valve Academic Research Consortium (6). The following events were evaluated: device success, vascular complications, bleeding, acute kidney injury (AKI), stroke, myocardial infarction, death, prosthetic valve performance, and the combined safety endpoint at 30 days. Vascular complications were divided into minor or major. Bleeding complications were divided into life-threatening or disabling bleeding and major bleeding. The AKI classification adopted the modified RIFLE (Risk, Injury, Failure, Loss, and End-stage kidney disease) criteria and includes 3 stages. Renal failure at baseline was defined as a creatinine clearance <60 ml/min. Stroke was further divided into major and minor. Prosthesis performance was evaluated at 1 month by quantifying transaortic gradient and aortic regurgitation (AR). Aortic regurgitation was graded as being absent (0), mild (1+), moderate (2+), and severe (3 to 4+) (7,8).
All the patients have been evaluated by a multidisciplinary heart team. Pre-procedural screening consisted of: transthoracic and/or transesophageal echocardiography, multislice computed tomography, and when required, coronary angiography. Transthoracic echocardiography was also performed at 24 to 48 h after procedure and at 1 month, to assess prosthesis performance and left ventricular function.
The indication for TAVI was based on the presence of symptomatic severe aortic stenosis (aortic valve area <1 cm2 or mean gradient >40 mm Hg) and high surgical risk features defined as a logistic European System for Cardiac Operative Risk Evaluation (logistic-European System for Cardiac Operative Risk Evaluation) ≥20%, Society of Thoracic Surgeons Predicted Risk of Mortality ≥10%, or associated comorbidities not captured by the 2 scores (e.g., prior thoracic radiotherapy, porcelain aorta, liver cirrhosis, marked patient frailty) (9). There were also some patients with symptomatic isolated severe AR, because of bioprosthetic failure who were at high risk for reoperation. The suitability for the transfemoral approach was determined by evaluating the ilio-femoral artery diameters on multislice computed tomography. A minimal lumen diameter (MLD) of 7 mm was necessary for the 22-F and 8 mm for the 24-F Retroflex delivery system but 6 mm for the Novaflex 18-F and 6.5 mm for the 19-F in the absence of circumferential calcifications.
All patients provided written informed consent for the procedures and subsequent data collection and analysis for research purposes.
The technical aspects of the procedure have been described elsewhere by our group (2,10). The ESV is a bovine pericardial leaflet valve mounted on a stainless steel stent. The new SXT has modified scallop-shaped leaflets that provide superior hemodynamic properties and coaptation. It is mounted on a cobalt-chromium alloy stent with thinner struts, thus, giving it a lower profile. Both valves have 2 sizes: 23 and 26 mm. The RetroFlex 3 is a balloon-expandable transcatheter delivery system where the valve is crimped directly on the balloon. It allows introduction of the 23- or 26-mm ESV valve with 22-F (outer diameter: 8.4 mm) and 24-F sheaths (outer diameter: 9.2 mm). The NovaFlex delivery system allows a reduction in sheath size to 18-F (outer diameter: 7.2 mm) for the 23-mm SXT valve and 19-F (outer diameter: 7.5 mm) for the 26-mm valve. The valve is crimped on the catheter proximally to the inflatable balloon and is mounted on the balloon in the descending aorta. The characteristics of the devices are illustrated in Figure 1.
Continuous variables are presented as mean ± SD, and categorical variables are presented as frequencies (%). Normality of the distribution of continuous variables was tested by means of the Kolmogorov-Smirnov goodness-of-fit test. Continuous variables were compared with the Student t test or Mann-Whitney U test where appropriate. Categorical variables were compared with chi-square statistic or Fisher exact test where appropriate.
Exploratory multivariable analysis was performed to assess the predictors of major vascular events (MVE) by means of logistic regression analysis with purposeful selection of covariates and simultaneous entry (11). Candidate variables included covariates associated at univariate analysis with MVE (all with a p value <0.1) as well as variables judged to be of clinical importance from previously published data. The results are reported as adjusted odds ratios (OR) with associated confidence intervals (CIs). The discrimination and calibration ability of the multivariable model was assessed by means of the C-statistic and the Hosmer-Lemeshow statistic (12). A p value <0.05 was considered statistically significant, and all reported p values are 2-sided. Statistical analysis was performed with SPSS software (version 16.0, SPSS, Inc., Chicago, Illinois).
Baseline and procedural characteristics
During the study period, 249 patients underwent TAVI at our center: 104 patients were treated with a Medtronic CoreValve (Medtronic, Inc., Minneapolis, Minnesota), and 145 were treated with an Edwards prosthesis. Of the patients who underwent implantation of an Edwards prosthesis, 25 were excluded, because the approach was transapical (n = 21) or transaxillary (n = 4). Thus, the current study cohort comprised 120 patients who underwent transfemoral implantation of an Edwards prosthesis: 54 with SXT between April and December 2010, and 66 with ESV from November 2007 to April 2010. Baseline clinical characteristics are presented in Table 1. Fifty percent of the patients were male, and the mean age was 80 ± 8 years. In 9 patients, the indication for TAVI was a degenerated aortic bioprosthesis, and 4 of them had isolated severe regurgitation. The patients treated with the SXT had a smaller MLD of the ilio-femoral arteries at the therapeutic access site than those treated with the ESV (7.27 ± 1.09 mm vs. 7.94 ± 1.08 mm; p = 0.002).
Procedural characteristics are shown in Table 2. The ratio between the outer diameter of the sheath size and MLD of the vessel was significantly lower in the SXT group (1.03 ± 0.15 vs. 1.13 ± 0.14, p = 0.001). Sedation and local anesthesia were used in 76% of cases. However, general anesthesia was used more frequently in the ESV group during the initial part of our TAVI experience (37.9% vs. 5.6%, p < 0.001). Percutaneous closure with a Prostar device (Abbott Vascular, Abbott, Park, Illinois) facilitated by the cross-over technique described by our group (10) was used more frequently in SXT than in ESV cases (98.1% vs. 83.3%, p = 0.008). However, this discrepancy was expected, because at the beginning of our experience we preferred surgical closure of the femoral artery in patients with borderline vascular access. In addition, in 4 patients who underwent ESV implantation, a vascular complication precluded the possibility of using the pre-closure device.
Periprocedural and 30-day clinical outcomes
The clinical outcomes are shown in Table 3. The rate of device success was high in both the SXT and ESV groups (96.3% vs. 92.4%, p = 0.45). The overall rate of MVE was 23.3%, but it was 3-fold lower in the SXT group (11.1% vs. 33.3%, relative risk: 0.40, 95% CI: 0.28 to 0.57; p = 0.004). In Online Table 1, we have described in detail the MVE that occurred and their management. The rate of MVE in patients who underwent percutaneous closure versus surgical closure was (9.6% vs. 0%, p = 1.00) in the SXT group and (34.4% vs. 20.0%, p = 0.87) in the ESV group. The minor vascular events were similar between the groups (25.9% vs. 30.3%, p = 0.59), and 18 of 34 (52.9%) of these events were due to Prostar failure, which occurred with a similar frequency in both groups (11.1% in SXT vs. 15% in ESV, p = 0.09). We observed 1 case of thoracic aortic dissection after ESV implantation and 1 case of annulus rupture and aortic dissection after SXT. The first underwent emergency cardiac surgery but died at 46 days. The second was treated by implanting another SXT to obtain hemostasis, but this was unsuccessful and the patient died in the catheterization laboratory. Life-threatening bleeding (18.5% vs. 27.3%, p = 0.25) and major bleeding (35.2% vs. 40.9%, p = 0.52) were not significantly different between groups. However, the need for blood transfusion showed a trend for a lower rate in the SXT group (33.3% vs. 57.6%; p = 0.08). Four cases of cardiac tamponade were observed in the SXT group, and 1 was observed in the ESV group. The cardiac tamponades were treated by pericardiocentesis in 3 cases, but 2 underwent cardiac surgery because of aortic dissection (as described in the preceding text) or rupture of the right atrial wall. The rate of AKI-stage 3 was 0 in the SXT group and 4.8% in the ESV group (p = 0.67). We observed only 2 cases of transient ischemic attacks and 2 cases of major stroke. No periprocedural myocardial infarction was observed. The rate of permanent pacemaker (PPM) implantation was 7.1%, without a difference between the groups.
The 30-day mortality was 1.7% (n = 2). Both patients died after SXT implantation. One died some hours after the procedure because of right ventricular failure; the other died in the catheterization laboratory because of annular and aortic dissection resulting in cardiac tamponade (described in the preceding text). Furthermore, 1 patient died in hospital at 46 days after emergency surgery for a post-implantation ascending aortic dissection (described in the preceding text).
The 30-day combined safety endpoint was lower in the SXT group (20.4% vs. 45.5%, relative risk: 0.44, 95% CI: 0.24 to 0.80; p = 0.004), mainly due to the lower rate of MVE in this group.
An echocardiographic evaluation was performed at 30 days in 88% of patients. The mean transaortic gradient was approximately 10 mm Hg in both groups. The AR was mild-to-moderate in 70.2% of SXT and 76.3% in ESV (p = 0.48). Two patients had a mean gradient >20 mm Hg: 1 after ESV implantation, and 1 after SXT implantation in a degenerated bioprosthesis. Four patients had severe AR at 1 month after ESV implantation (Table 4).
Univariate and multivariable analyses
Results of the univariate and multivariable analyses are presented in Online Table 2. The univariate analysis identified 4 potential predictors for MVE: SXT implantation (SXT vs. ESV; OR: 0.25, 95% CI: 0.09 to 0.67; p = 0.004), ratio of sheath outer diameter/MLD (OR: 18.70, 95% CI: 1.04 to 336.41; p = 0.04), body surface area (OR: 0.05, 95% CI: 0 to 0.80; p = 0.03), and female sex (OR: 2.14, 95% CI: 0.89 to 5.14; p = 0.08). On the multivariable analyses, the SXT (OR: 0.25, 95% CI: 0.09 to 0.67; p = 0.006) emerged as the only significant independent predictor of reduced MVE, whereas female sex showed a trend for increased MVE (OR: 2.16, 95% CI: 0.87 to 5.35; p = 0.09). The Hosmer-Lemeshow goodness-of-fit test probability value was 0.99, confirming good calibration and fit of the multivariable model. The C-statistic for the final model was 0.7, indicating acceptable discrimination.
The main findings of this study are: 1) introduction of the SXT valve and NovaFlex delivery allowed the performance of transfemoral TAVI in patients with more severe peripheral artery disease with a reduction of MVE; 2) transfemoral TAVI with both the SXT and ESV was associated with a high device success rate and relatively low periprocedural and short-term outcomes in high-risk elderly patients; and 3) the SXT seems to have the same short-term performance as the ESV.
The transfemoral route is at the present time limited by the large diameters of the delivery sheaths required and the inadequate diameters of the ilio-femoral vessels to accommodate these sheaths. Because the prevalence of peripheral arterial disease is high in this population, many patients are not suitable for transfemoral TAVI. To avoid this problem, companies are working to produce devices with smaller sizes that will make this procedure a viable option in more patients; at the same time, they could reduce the rate of vascular complications, which are the most frequent cause of morbidity after TAVI. Until recently, there has not been a uniform definition of post-TAVI events in the published data, but recently Leon et al. (6) published the new Valve Academic Research Consortium endpoint definitions. As can be seen in Online Table 3, there is a wide variation (8% to 16%) in the incidence of MVE in transfemoral TAVI with the Sapien valves (1,3–5,13–16).
In our study, the incidence of total MVE was 23.3%. However, due to the lack of a standardized definition for vascular complications and the different definitions used in previous studies, it is difficult to make any direct comparisons. We observed a significantly lower rate of MVE in patients treated with SXT compared with ESV (11.1% vs. 33.3%, p = 0.004). Moreover, we found that the ratio of the sheath outer diameter size to vessel MLD was a univariate predictor of MVE (1.14 ± 0.15 vs. 1.07 ± 0.15, p = 0.04). Furthermore, this ratio was lower in the SXT group than in the ESV group (1.03 ± 0.15 vs. 1.13 ± 0.14, p = 0.001). This suggests that vascular injury increases as the operator attempts the procedure with diseased peripheral vessels that have an inadequate size for the sheaths. Thus, the SXT might have a broader clinical application, because it allows TAVI in patients with more severe peripheral vascular disease by reducing the ratio of sheath outer diameter size to vessel MLD but, at the same time, lowering the risk of MVE. The lower MVE risk with SXT was confirmed on multivariable analysis.
Bleeding is an important safety endpoint in evaluating the TAVI procedure, but similar to MVE, the definitions of bleeding are variable (1,13,16). In our study, the incidence of life-threatening bleeding and major bleeding seems not to differ significantly between the groups, but we observed a trend for a lower need for blood transfusion in the SXT group (33.3% vs. 57.6%). Interestingly, there was a trend of more cases of cardiac tamponade in the SXT group. We believe that this complication could not be attributed to the new delivery system or to changes in the technical aspects of the procedure but to the recent introduction and training of new operators in our team.
However, even if the MVE were more frequent in the ESV group, we observed only 3 in-hospital deaths (2.5%) and a low rate of 30-day mortality (1.7%), without a significant difference between the 2 groups. As discussed in the preceding text, only 2 patients died because of vascular complications. We believe that the low rate of early mortality in our population, despite the high rate of vascular events, is due to the ability of our operators to promptly treat such complications. In addition, the use of a lower-diameter sheath decreased the occurrence of vascular complications. We think that expertise of TAVI operators in the percutaneous treatment of vascular complications is paramount for reducing mortality in this high-risk patient group.
As already described in published data, the rate of PPM is low after Edwards Sapien valve implantation compared with the CoreValve (Medtronic) (16). In this study we didn't find any difference in the need of PPM in the ESV and SXT groups, and we have confirmed a low rate in both groups (approximately 8%).
Finally, this study demonstrated that the SXT has similar performance at 1-month compared with the ESV. The rate of device success was high with both valves, and the grade of AR in both groups was mainly mild or moderate; meanwhile, the mean transaortic gradient was approximately 10 mm Hg.
The main limitation of this study is its retrospective observational design and lack of randomization. However, when the SXT was introduced in Europe, the ESV became no longer commercially available. We think that these data are important and might provide useful information with regard to sample size, should the U.S. Food and Drug Administration require a randomized trial before approval of the new SXT valve. Another limitation could be the different level of experience of the operators through the study period.
TAVI with an Edwards prosthesis was associated with high implantation success and low early mortality in patients at high-risk for surgery, with the 2 prostheses having a similar short-term performance. The requirement for a smaller ilio-femoral diameter and lower risk of major vascular complications extended the clinical applications of the SXT system.
For supplementary tables, please see the online version of this article.
The authors have reported that they have no relationships to disclose. Drs. Mussardo and Latib contributed equally to the manuscript and are joint first authors.
- Abbreviations and Acronyms
- acute kidney injury
- aortic regurgitation
- confidence interval
- Edwards Sapien THV
- minimal lumen diameter
- major vascular event(s)
- odds ratio
- permanent pacemaker
- Sapien XT THV
- transcatheter aortic valve implantation
- Received February 6, 2011.
- Revision received May 2, 2011.
- Accepted May 6, 2011.
- American College of Cardiology Foundation
- Godino C.,
- Maisano F.,
- Montorfano M.,
- et al.
- Rodes-Cabau J.,
- Webb J.G.,
- Cheung A.,
- et al.
- Thomas M.,
- Schymik G.,
- Walther T.,
- et al.
- Webb J.G.,
- Pasupati S.,
- Humphries K.,
- et al.
- Leon M.B.,
- Piazza N.,
- Nikolsky E.,
- et al.
- Perry G.J.,
- Helmcke F.,
- Nanda N.C.,
- Byard C.,
- Soto B.
- Zoghbi W.A.,
- Chambers J.B.,
- Dumesnil J.G.,
- et al.
- Fried L.P.,
- Tangen C.M.,
- Walston J.,
- et al.
- Hosmer D.,
- Lemeshow S.
- Webb J.G.,
- Altwegg L.,
- Boone R.H.,
- et al.
- Himbert D.,
- Descoutures F.,
- Al-Attar N.,
- et al.
- Gurvitch R.,
- Wood D.A.,
- Tay E.L.,
- et al.