Reed Miller

May 18, 2012

May 18, 2012 (Paris, France) — Early data from the SOURCE XT registry on high-risk patients treated with the Sapien XT (Edwards Lifesciences) transcatheter aortic valve show that this next-generation TAVI device is safe with either the transapical- or transfemoral-access approach.

One-month data from 2706 patients in SOURCE XT were the subject of several presentations at EuroPCR 2012. Mortality in the 1694 patients implanted with the Sapien XT via the transfemoral approach was 4.3%, among the lowest rates ever reported in a registry of transfemoral TAVI patients, Dr Olaf Wendler (King's College Hospital, London, UK) reported. The one-month mortality for the 906 patients implanted with the Sapien XT via the transapical approach was 9.9%, and the mortality rate for all of the non–transfemoral-access routes combined — there were also 98 transaortic and eight subclavian patients — was 9.7%.

In an interview with heartwire , Wendler emphasized that paravalvular leak was low in both the transapical and transfemoral groups. There was no paravalvular leak or only a trace of a leak in nearly 65% of the transfemoral group and 79% of the transapical group. Moderate to severe leaks were seen in only 5.8% and 3.5% of the transfemoral and transapical groups, respectively. "Knowing that paravalvular leakage is a predictor of mortality in the long term, that is a very positive message from this valve," he said. "With this excellent performance and low incidence of paravalvular leakage, we would expect to see even better outcomes [than we've seen with older models of the valve] at the one-year [follow-up]."

More insight into the transapical approach

Mortality rates were lower for patients treated with the transfemoral approach than with the transapical approach because the patients chosen by the heart team for the other approaches were generally sicker to begin with, the results show. The transapical patients had an average logistic EuroSCORE of almost 22, compared with 20 for the transfemoral patients (p=0.0004), and were more likely to have a EuroSCORE over 40 (10.8% vs 6.7%). In patients with a EuroSCORE of 40 or below, 30-day mortality was 4% in transfemoral patients and 8.4% in transapical patients, but in patients with EuroSCORE over 40, the mortality rates were 7% and 22.4% for the transfemoral and transapical groups, respectively. Compared with the transfemoral patients, the transapical patients had a lower mean left ventricular ejection fraction and were significantly more likely to have had coronary disease, a previous MI, atrial fibrillation, and/or a previous aortic bioprosthesis.

The decision of which access route to use was made by the heart team at each of the 94 centers in the study. Each team included an interventionalist and a cardiac surgeon as primary investigators. Usually, if the transfemoral route was available, that's the route they chose, but some patients had too much peripheral disease or other complications that made the transfemoral approach too difficult, Wendler said.

Wendler pointed out that for transapical patients getting the largest Sapien XT, the 29-mm model, mortality was only 6%, as low as any ever seen for the transapical approach in a clinical registry, he said. The 29-mm valve was not available for transfemoral patients at the time of this study. Therefore, "we can say the 29-mm group is the one group where the outcomes reflect what the outcomes would be if patients both with and without peripheral vascular access were treated only with transapical access. And because the mortality is so low . . . it shows us that the reasons that transapical outcomes are worse than transfemoral outcomes is the fact that patients with good peripheral access also often have lower risk and therefore have better outcomes," he says. "It confirms the hypothesis that the mortality that we see in the whole transapical cohort is based on the higher risk of the patients and not the procedure."

Given that the transapical results look worse because of baseline differences rather than the procedure itself, it is still possible that the transapical route may prove to be better overall for some patients eligible for either option, Wendler said. "The hypothesis, based on what data we have, is that transfemoral is less traumatic than transapical, and that's why, although we don't have hard data on this, by making the decision for transfemoral first, one is not doing something wrong," he said. "The transfemoral-first approach is working, but we don't have hard scientific data to say that transfemoral is better than transapical in patients where both access modes are feasible."

Wendler also observed that although major vascular complications were more common after the transfemoral procedure, bleeding was more common with the transapical procedure. With either approach, "procedural complications were rare and early hemodynamic function excellent using the Sapien XT," he said.

Also, left ventricular ejection fraction improved significantly in both the transapical (53.2% to 54.9%) and transfemoral groups (55.3% to 58.2%). "In the past, it's been mentioned that because the valve is implanted through the left ventricular muscle [in the transapical approach], it must have a negative impact on the left ventricular function of the heart," Wendler said. "But our data show nicely that the ejection fraction improves after the procedure as significantly after transapical implantation as it does after transfemoral implantation. There is no indicator that the transapical access has a negative impact on left ventricular performance after the procedure."

New design is more stable

The Sapien TAVI valve, available in the US and Europe, is bovine pericardial tissue on a stainless-steel frame. The Sapien XT is bovine pericardial tissue on a cobalt-chromium frame. It also features semiclosed leaflets and a smaller crimped profile than the Sapien and comes in 23-mm, 26-mm, and 29-mm diameters, whereas the Sapien is available only in 23-mm and 26-mm diameters.

"The most important difference [between Sapien XT and Sapien] is that the valve is much more stable during implantation," Wendler said. "The way one can see that in our data set is in procedural complications; although procedural complications were low in the previous SOURCE registry [studying the Sapien], they are even lower in SOURCE XT." Rates of conversion to surgery, cardiac tamponade, annular dissection, coronary occlusion, and valve embolization were all below 1% in SOURCE XT for either access route. "The [reduction] in complications is nicely explained by the more stable position of the valve during deployment."

Edwards recently announced that the 29-mm version of the transfemoral Sapien XT has earned a CE Mark for commercialization in Europe. The Sapien XT and the NovaFlex transfemoral and Ascendra 2 transapical delivery systems received CE Mark approval in March 2010.

Wendler has research contracts with Edwards Lifesciences and consults for Edwards, JenaValve, and St Jude Medical.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.