PARTNER 2: New Sapien XT Matches First TAVR Device, With Lower Complications

Shelley Wood

March 10, 2013

SAN FRANCISCO, California — A lower-profile, next-generation version of the Sapien transcatheter aortic-valve system known as the Sapien XT (Edwards Lifesciences) was associated with similar rates of death and stroke as the first-generation device, results from the PARTNER 2 trial show. The device, tested here in inoperable patients ("cohort B"), also appeared to improve on several procedural aspects of valve deployment, including lower anesthesia time and fewer vascular complications.

Dr Martin Leon

Dr Martin Leon (Columbia University, New York, NY), presenting the results here during the second late-breaking clinical-trial session of the American College of Cardiology 2013 Scientific Sessions, highlighted several characteristics of the new device, including an improved frame geometry, less metal content, and a lower crimp profile. Perhaps most important, the device can be delivered via a lower-profile sheath. For example, said Leon, a 23-mm valve can now be delivered via a sheath with a 33% reduction in the sheath cross-section.

PARTNER 2, a trial requested by the FDA, was undertaken to see whether these "important enhancements," as Leon called them, were associated with equivalent or improved clinical outcomes as compared with the first approved device.

What's New Is New Again

PARTNER 2 randomized 560 patients who were ineligible for surgery to transfemoral transcatheter aortic-valve replacement (TAVR) using the new Sapien XT or transfemoral TAVR using the original Sapien device. All patients were first reviewed by a conference call involving both a surgeon and an interventionalist, among others. Mean age of the patients was 84, and mean baseline STS scores were between 10 and 11 in the two groups.

Procedures for both devices took roughly the same amount of time, just over 100 minutes, but anesthesia time was significantly lower in the Sapien XT group, by about 15 minutes. Aborted procedures were also less common: two vs eight, in favor of the new device (p=0.06).

The primary end point--all-cause death, disabling stroke, and rehospitalization--was similar between the two devices, as were rates of individual clinical events at both 30 days and one year. Mortality at one year was roughly 23% in both arms, but of note, Leon said, rates of all-cause mortality at one year were lower for both arms in PARTNER 2 than they were for the inoperable, TAVR-treated cohort in the original PARTNERtrial, at 30.7%.

Stroke occurred in approximately 4.5% of patients in both groups in PARTNER 2--that's down from a rate of 7.8% seen for the Sapien in the original PARTNER B trial, reported by heartwire in 2010. In a morning press conference, Leon noted that the stroke rates in PARTNER 2 were similar to what's now seen in the continued-access cohort from the original PARTNER trial.

Major vascular events were lower for the newer, lower-profile device than for the first-generation Sapien, with fewer perforations and dissections.

Thirty-Day and One-Year Outcomes: Sapien vs Sapien XT

End point Sapien (%) Sapien XT (%) p
30-d all-cause death, disabling stroke, and rehospitalizations 15.3 17.0 0.60
Major vascular events, 30 d 15.5 9.6 0.04
Perforations 4.8 0.4 0.003
Dissections 9.2 4.3 0.03
All-cause mortality, 12 mo 23.7 22.5 NS
Stroke, 12 mo 4.6 4.5 NS


The only worrying end point to receive attention here today was a nonsignificant trend toward more moderate and severe paravalvular regurgitation in the newer device.

"We're still trying to understand a little bit more about that trend," Leon told reporters in a press conference. One "confounding" issue, he noted, might be the higher use of second and even third valves in the course of a single procedure to get optimal valve placement, which was more common with Sapien than Sapien XT and which tends to also resolve some if not all paravalvular regurgitation. Leon also noted that the next-next generation Sapien device, which he called "Sapien 3," boasts an external cuff, which has "eliminated, or all but eliminated, paravalvular leak . . . so you may have to wait another generation of devices to solve that problem."

Commenting on PARTNER 2 for heartwire , Dr John Webb (St Paul's Hospital, Vancouver, BC) pointed out that the greater degree of difficulty operators faced in implanting the original Sapien device made malpositions more common, leading to the use of second valves. "The difference was probably an artifact of this, [and] in any event the difference was not significant."

He predicts that real-world results may be even better. "The difference between the large- and small-profile devices will be much more dramatic in the large number of patients with small arteries poorly suited to large sheaths."

Also asked for his views, cardiovascular surgeon Dr Tim Gardner (Christiana Care Health System, Newark, DE) said he doubted the paravalvular leaks "are going to amount to anything.

"I actually think that the FDA may be too compulsive in requiring a [randomized clinical trial] for incremental device improvements like those that have been incorporated in the Sapien XT. I know that there have been issues with some EP device and equipment modifications, and there was the infamous Bjork-Shiley valve manufacturing modification that led to prosthesis failures.  However, the Sapien XT is an appropriate improvement on the original device, it's been implanted extensively elsewhere, including in Canada, I believe, and I would expect it to do just what they showed--fewer insertion problems and equal efficacy."

"A Worthwhile Advance"

Overall, Leon concluded, "The new lower-profile Sapien XT system was associated with improved procedural outcomes and similar low 30-day morality and strokes, as well as reduced vascular complications and similar one-year major clinical events and valve performance" in inoperable aortic-stenosis patients.

"Sapien XT represents a worthwhile advance with incremental clinical value and is the preferred balloon-expandable transcatheter heart-valve system," Leon said.

In a panel discussion following the PARTNER 2 presentation, interventionalists tried to tease out how much the better procedural and vascular outcomes in this trial were related to the device changes as compared with factors such as improved operator experience/technique and better patient selection.

Leon responded, noting that patient selection for PARTNER 2 had actually been very similar to that of PARTNER, but that without question, "we've learned something" over the years. "We do less trauma to the native valve, we watch the patients better, and we pay more attention to adjunct pharmacotherapy."

Dr Roxana Mehran (Mount Sinai School of Medicine, New York, NY), one of the panel members, pointed out that adjunctive pharmacotherapy is really the next frontier for improving on TAVR generally. "This is a place where we still have a ways to go," she noted.

Edwards Lifesciences sponsored PARTNER 2. Leon disclosed receiving grant/research support from Abbott, Boston Scientific, Edwards, and Medtronic and holding major stock/shareholder/equity with Sadra, Claret, Valve Medical, and Apica.


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.
Post as: