Published 2-Year Partner 3 Data Shed New Light on TAVI, SAVR Choice in Low-risk Patients

Martin Leon cautions against “overinterpreting” the numbers, since the full low-risk TAVI story will take years to complete.

Published 2-Year Partner 3 Data Shed New Light on TAVI, SAVR Choice in Low-risk Patients

Published yesterday, 2-year results from the PARTNER 3 trial should help to flesh out the bigger picture of TAVI versus surgery in low-risk patients, but the data can’t yet answer the question of which procedure is the best choice in this group.

The results, originally released last year at the virtual 2020 American College of Cardiology (ACC) meeting and now published in the March 9, 2021, issue of the Journal of the American College of Cardiology, show a narrowing lead for TAVI regarding the combined primary endpoint of death, stroke, and cardiac rehospitalization over surgical aortic valve replacement at 2 years (11.5% vs 17.4%; HR 0.63; 95% CI 0.45-0.88) as compared with the 1-year results, where the TAVI showed superiority.

However, lead author Martin B. Leon, MD (Columbia University Irving Medical Center/NewYork Presbyterian Hospital, New York, NY), cautioned against “overinterpreting” these data given that the event rates “are in the low single digits” and follow-up is planned out to 10 years. “It emphasizes that follow-up is important and the whole story of a trial like this will be told over the course of many years, not just the first year where we present the primary endpoint outcomes,” he told TCTMD. “If the event-rate lines cross and suddenly [the rate for] surgery is now less than TAVR in year 3 or 4, to me that becomes meaningful. But there's at least as high a likelihood that they will remain flat or parallel in subsequent years.”

Leon also stressed the fact that PARTNER 3 was not an all-comers trial and hence the results cannot be broadly applied. “This study really applies to the population that we included in the trial—we didn't include bicuspids, we didn't include very young patients, and we didn't include a lot of patients that are in the low-risk category,” he said. “People have to understand: this was a low-risk trial where we felt there was the best chance of achieving good results with TAVR, and we didn't treat very complex anatomies so we should be a little careful in generalizing this to the entire low-risk population.”

Another advantage of follow-up studies is their ability to look at secondary endpoints more closely, Leon added, citing more late valve thrombosis with TAVI than surgery (2.6% vs 0.7%; P = 0.02) at 24 months “that we need to continue to follow.” On the other hand, PARTNER 3 is the first trial to show a sustained quality-of-life benefit for TAVI over surgery at 2 years (P = 0.002), he said.

Lastly, Leon cited the new VARC 3 valve durability definitions used in the study as a major strength. “This is the appropriate standard definition that should involve bioprosthetic valves, either surgical or transcatheter,” he said. “Looking at the serial echocardiograms and integrating it with clinical data at least out to 2 years, there was no difference between the transcatheter and surgical bioprosthetic valves. Thematically, that's the way that we need to be analyzing these data in the future if we're going to be able to determine the subtle differences that might be clinically important.”

Three-year outcomes from the trial will likely be presented during the TCT meeting this fall, Leon said.

Digging Into Secondary Endpoints

In an accompanying editorial, Bernard Prendergast, MD (St Thomas’ Hospital, London, England), and colleagues highlight that four of the seven deaths observed in the TAVI group were noncardiac, while just three of the six strokes were disabling and confirmed by cerebral imaging.

Those details caught the eye of Kendra Grubb, MD (Emory University, Atlanta, GA), who said she didn’t especially realize the nature of the deaths in the study when the results were first released last year.

“Although intracranial hemorrhage is associated with a stroke that was due to the TAVR—[so] it could possibly be attributed to the valve—certainly a traumatic event, cancer, suicide, and sepsis are less likely to be the result of the TAVR,” she told TCTMD. “When Dr. Mack presented this data, I did not appreciate that fact. It really came across in his presentation that the driver of the difference between TAVR and SAVR was rehospitalization and listening to the data be presented, it seemed like there was an excess of deaths between 1 and 2 years in the TAVR arm. With this knowledge that really they're not cardiac deaths, that does kind of change my understanding or change my threshold for these TAVR valves.”

Additionally, while Grubb said the quality-of-life findings are exciting, she urged caution in interpreting them, because it’s not yet known if there’s a trade-off for that gain. “Presumably, if the valve has same durability, then that early quality-of-life increase is really critical. But if instead we find that, for whatever reason, the TAVR valves fail earlier, in my young patient I don't necessarily believe that that quality-of-life increase at 2 years would be worth a durability price to pay,” she noted.

According to the editorialists, the echocardiographic results showing no connection between outcomes and structural valve deterioration, bioprosthetic valve failure, or moderate/severe paravalvular leak “confirm the early durability of current TAVR devices and the combined advantages of external sealing cuffs incorporated into contemporary transcatheter heart valve designs coupled with accurate CT annular sizing.”

Even so, many questions about transcatheter valve durability remain. Grubb said the next pieces that need to be understood are the impact of the higher valve mean gradient and lower effective orifice area (EOA) observed with Sapien valves as well as the effect of valve thrombosis. “Are we going to see these valves fail earlier because of those things?” she asked. “I certainly think that higher mean gradients after TAVR versus SAVR cause me to pause. And even though EOAs were similar, what does that higher gradient at 2 years translate to at 10 years? Then you add the higher thrombosis rate, and there are a lot of unknowns.”

Because many of the instances of valve thrombosis were only found when patients underwent echocardiogram-driven CT studies, and not because of symptoms, treatment is often successful, Grubb said. “But the question is, should we be following all TAVR patients with the same level of scrutiny as we did in these trials? I certainly think yes. That if we have a patient on their annual echo and the gradient is higher than what we expect, then we should be doing a follow-up CT, we should be anticoagulating them and trying to maximize these valves assuming that a higher gradient would lead to worse durability. I have no proof for that yet, but I believe that that's probably the right way to go.”

Leon agreed that more work needs to be done comparing transcatheter valve durability with that of surgical devices.

“We're not seeing any early signals certainly up to 5 years, and the question now that we begin to follow patients out to between 5 and 10 years is: will we begin to see any important differences?” he said. “Now surgical valves that have been shown to be less than optimally durable have usually shown some evidence or signals of lack of durability that begin to occur in the first 5 years but then accelerate dramatically between 5 and 10 years. So in the next several years, we'll have enough patients with enough follow-up to know if these bioprosthetic valves have a similar durability as surgical valves. And that's an important consideration in terms of making lifelong decisions for patients, particularly younger patients.”

At this point, there seems to be agreement that TAVI is the optimal treatment for patients over 80 while surgery should remain the default for patients younger than 70, write Prendergast and colleagues. As for those in between, “the choice should be determined following a shared decision-making process incorporating the pros and cons of SAVR and TAVR guided by patient-specific anatomic and clinical considerations, the potential need for repeat valve interventions over a lifetime journey (including their mode, sequence, and associated risks), and the patient’s values and preferences.” That decision-making process is in line with the new ACC/American Heart Association valvular heart disease guidelines.


Note: Leon and several other co-authors are faculty members of the Cardiovascular Research Foundation, the publisher of TCTMD.

  • The PARTNER 3 Trial was funded by Edwards Lifesciences.
  • Leon reports receiving grant support, paid to his institution, from Edwards Lifesciences, Medtronic, Abbott, and Boston Scientific; and has received advisory board fees from Medtronic, Abbott, Boston Scientific, Gore, and Meril Life Sciences.
  • Prendergast reports receiving unrestricted research and educational grants from Edwards Lifesciences and receiving speaker/consultancy fees from Abbott, Anteris, and Edwards Lifesciences.
  • Grubb reports serving as a speaker, proctor, and PI for Edwards Lifesciences and Medtronic.