March 31st, 2014

Are We Equipped to Make a CHOICE Between TAVR Devices?

CardioExchange’s Rick Lange and David Hillis ask E. Murat Tuzcu for his perspective on the CHOICE trial comparing the CoreValve and Edwards Sapien XT devices for transcatheter aortic valve replacement (TAVR). Dr. Tuzcu is a coauthor of the JAMA editorial on CHOICE. For CardioExchange’s news coverage of CHOICE, click here.


The Comparison of Transcatheter Heart Valves in High-Risk Patients With Severe Aortic Stenosis: Medtronic CoreValve vs Edwards SAPIEN XT (CHOICE) study is the first randomized clinical trial to compare two different TAVR heart-valve technologies. The participants were 241 high-risk patients with severe aortic stenosis in Germany.

The rate of “device success,” the primary endpoint, was significantly higher with the balloon-expandable Sapien XT device than the self-expandable CoreValve (95.9% vs. 77.5%). That finding was attributable to a significantly lower frequency of more-than-mild paravalvular aortic regurgitation with the Sapien XT device (4.1% vs. 18.3%) and the Sapien group’s significantly lower rate of need for implanting 2 devices to achieve an acceptable hemodynamic outcome (0.8% vs. 5.8%).

Major adverse cardiovascular and cerebrovascular events at 30 days were numerically, but not statistically, higher in the Sapien group (6.6% vs. 3.4%; RR, 1.93; 95% CI, 0.60­–6.25; P=0.38). The cardiovascular mortality rate at 30 days was similar in the two groups (4.1% and 4.3%, respectively).


Lange and Hillis: This relatively small study revealed no difference between the devices in 30-day cardiovascular mortality. Are the primary-endpoint differences clinically meaningful?

Tuzcu: Previous studies have shown a positive correlation between moderate and severe aortic regurgitation (AR) and mid-term mortality. In the PARTNER trial, even mild AR correlated with 1-year mortality. Thus, the differences in the primary endpoint in the CHOICE trial should not be taken lightly. We published a meta-analysis (of 45 studies) that also showed a higher rate of post-TAVR AR with the self-expanding (SE) CoreValve than with the balloon-expandable (BE) Edwards valve. In the search for an ideal transcatheter valve, elimination of post-TAVR AR is a critical step that cannot be ignored.

Another important contributor to the device success, as reflected in the CHOICE trial’s primary endpoint, was the implantation of a second valve during the index procedure. Analyzing the relevant PARTNER trial data, Makkar et al. concluded that valve in valve implantation to treat acute aortic regurgitation predicts 1-year mortality. In summary, the primary-endpoint differences are clinically meaningful.

Lange and Hillis: Is the nonsignificant difference between the devices in the rate of major adverse cardiovascular and cerebrovascular events real? How do you weigh this against reduced incidences of paravalvular aortic regurgitation and need for a second valve with the balloon-expandable device?

Tuzcu: Major adverse cardiovascular and cerebrovascular events were 6.6% and 3.4% with the BE and SE valves, respectively, predominantly due to the statistically nonsignificant (P=0.33) difference between the stroke rates. We learned from presentation of the CHOICE trial at ACC.14 that there were 4 minor strokes in the BE group but none in the SE group. The major stroke rates were essentially same in the 2 groups.

This month we published a meta-analysis in JACC focusing on the post-TAVR stroke issue in 23 multicenter and 33 single-center studies. The stroke rate did not differ between the BE and SE valves. Data suggest that the stroke rate after TAVR decreased over time. This is likely due to better patient selection, improved technology, and enhanced experience.

Although I don’t see a cause for alarm in the stroke data from CHOICE and other studies, I think we should be vigilant for possible differences not only between the two valves tested in this trail, but also among the new-generation valves entering clinical use (because it is the most consequential complication of the TAVR procedure).

Lange and Hillis: If you needed to undergo TAVR today, do you have enough information to decide between the two devices?

Tuzcu: This is not an easy question to answer. I would acknowledge that I do not have conclusive information. But if I had to make a decision with the totality of information we have today, I would probably look at anatomical characteristics. If I had a small, heavily calcified annulus and LVOT, short distance between annulus and coronary ostia, I might choose an SE valve. This would minimize my risk for annular rupture and coronary obstruction, both of which are infrequent but potentially lethal complications. Although we don’t have a lot of data, if I had predominantly regurgitant and minimally calcified aortic valve disease, I would ask for an SE device. Lacking those characteristics, I would choose the BE valve, thinking that I would have a lower risk of needing a pacemaker, a greater chance of device success, and maybe less risk for rehospitalization.


Share your thoughts on Dr. Tuczu’s analysis of the findings from CHOICE.

To view all of our coverage from the ACC meeting, go to our ACC.14 Headquarters page.

One Response to “Are We Equipped to Make a CHOICE Between TAVR Devices?”

  1. Michael Reardon Reardon, MD says:

    Several things about your comments should be pointed out. First the Choice trial used angiography to estimated the degree of PVL which was the driver of success differences. This is in no way a reasonable standard and differed from their echo findings. The meta analysis that have shown ES to have a lower PVL rate that Corevalve are all based on registries only without core labs. We have several good prospective studies with echo core lab adjudication of PVL; PARTNER IA, PARTNER IB and PARTNER II B on the ES side and the Corevalve US Pivotal extreme risk and high risk studies. In all of these trials which are the only core lab adjudicated trials Corevalve had a lower PVL rate. You can not go to registries when good prospective core lab adjudicates studies exist. Choice does not tell us how long they waited after Corevalve implant to do an angiogram. Standard practice is to wait 10 minutes as we routinely seeing any PVL present get better over this time frame as the valve expands. Choice also did not use 3D imaging to choose valve sizes in all patients – this too is below any standard that should be used in contemporary practice. To make any conclusions based on Choice is a real stretch as is the use of registry data. You do have one valve, Corevalve, that has show superiority to sAVR in the high risk population. More importantly the survival curves in PARTNER A diverge nuptially with TAVR doing better but quickly converge. With the high risk Corevlave trial they diverge with TAVR doing better and never converge remaining apart and parallel at 2 years. This is likely due in my opinion to the improvement in early PVL that we see in 80% of Corevalve cases due to the continued expansion of the Nitinol frame. In PARTNER A and B this did not occur and in PARTNER II B it got worse with time
    The PARTNER trials and the Corevalve US Pivotal trials were not designed to test these valve against each other and I would not use them to comment on choice but Making a valve choice based on valve success from a poor study like Choice is not good advice.