TAVI Registries Offer Outcome Clues as Low-Risk Numbers Climb

The latest numbers from the Society of Thoracic Surgeons/American College of Cardiology TVT Registry show that TAVI outcomes are improving over time across a range of important metrics, including mortality and complications, but have remained static in other key areas, most notably stroke and pacemaker implantations.

Moreover, COVID-19 is leaving its mark on procedure volumes, with inevitable implications for patient care, said Joseph Bavaria, MD (University of Pennsylvania, Philadelphia), chair of the TVT Registry steering committee from 2017 to 2020, who presented what he called “hot from the press” data in a plenary session at the European Association for Cardio-Thoracic Surgery (EACTS) last weekend.

Meanwhile, new numbers from the German Aortic Valve Registry (GARY), dating back nearly 10 years and presented during the same session, may give pause—particularly for heart teams discussing the best options for low-risk patients.

In the United States, by federal mandate the TVT Registry captures every transcatheter aortic valve replacement performed and now represents the largest source of TAVI data in the world, Bavaria noted.

TAVI is being adopted very, very significantly across the country, and the number of sites has gone up dramatically. Joseph Bavaria

“I think the most interesting thing is that TAVI is being adopted very, very significantly across the country, and the number of sites has gone up dramatically,” Bavaria told TCTMD. “There could be an argument made as to whether the increase in number of sites since the [national coverage determination] went into effect has been exactly what was intended, which was to increase access in geographically disparate regions, but the fact is the number of sites is up.”

And even as TAVI has expanded to new sites, a longitudinal look at outcomes over the past 9 years points to steady gains.

“There are a lot of good outcomes: the mortality rates have steadily come down, the discharge to home has steadily come up, and the transfemoral cannulation has steadily gone up,” Bavaria said.  

More worrisome, however, is that both stroke and pacemaker rates have remained “static,” which should be top of mind for heart teams, especially when considering the options for younger patients, he said.

And whether the rising proportion of low-risk TAVI procedures makes sense given the kinds of numbers emerging in trials and registries is unknown. “I think it's going to be a really interesting couple of years regarding the low-risk patient population,” Bavaria said.

US Real-world Outcomes

According to the figures Bavaria showed at EACTS, median length of stay following TAVI has dropped from 4 to 6 days back in 2014, to just 1 day for low- and intermediate-risk patients in 2020. Vascular complications have followed a similar trajectory.

More importantly for patients, both 30-day and 1-year mortality have declined since the launch of the TVT Registry: from 7.5% and 26.4% in 2012 to 2.3% and 11.8% in 2020. Of note, 30-day mortality has remained low and decreased slowly over time; yet it did not decrease in 2019-2020, despite the influx of low-risk patients, Bavaria noted. Focusing solely on low-risk patients (median STS score 2), 1-year mortality is 5.3%, down from 15% in 2014.

Permanent pacemaker (PPM) implantation, however, has not enjoyed the same success and “this I think is most important—we’re not proud of this in the US,” said Bavaria, noting that he believes other parts of the world are in the same boat. “Despite newer-generation valves and lower-risk patients, the overall new pacemaker rate is essentially unchanged over 7 years,” he said during his talk. Even in low-risk patients, the PPM rate is 8%.

“The overall pacemaker rates are coming down just a touch, but they're probably not coming down as low as you would expect from the lower-risk patient profiles that are being done, and that's kind of a big concern,” he told TCTMD. “It’s a little disturbing that the overall pacemaker rates have only been inching down and are probably not inching down if adjudicated for the risk profile. . . . It makes you think that there might be a platform issue.”

A pacemaker “is not such a bad thing for an older person, but it is not a good thing for a younger person,” he continued. “So that's one area where I think the TVT steering committee believes that the US needs to do better.”

We really need to crack the pacemaker problem if we're going to treat low-risk patients universally. Bernard Prendergast

Trends in stroke rates are “also disturbing,” Bavaria said, presenting a graph showing largely parallel lines for in-hospital, 30-day, and 1-year stroke rates over time. In-hospital and 30-day stroke rates, which were at 2% and 2.8% back in 2012, still hover at 1.4% and 2.1% in 2020, with no marked changes following the FDA approval of cerebral protection devices in 2018. “Despite newer-generation valves and lower-risk patients, the overall new stroke rate is essentially unchanged over 7 years,” he said.

Bernard Prendergast, MD (Cleveland Clinic, London, England), who also was part of plenary session, agreed, calling pacemaker implants the “Achilles' heel” of TAVI today. “We really need to crack the pacemaker problem if we're going to treat low-risk patients universally,” he commented in an interview. “A pacemaker rate exceeding 5% is not acceptable in a young, low-risk patient.” Operators and manufacturers are focusing on new devices, new techniques, and new algorithms for patient selection in an effort to bring these rates down, said Prendergast. 

The stroke rates represent good and bad news, Bavaria stressed to TCTMD: “It's good that those stroke rates are pretty low, but we would hope that the stroke rates would actually trend down a little faster, again, especially since we're going into lower-risk patients.” More detailed analysis of these trends are needed, he added.

Moreover, one of the chief aims when the TVT Registry was created was “have and to utilize a quality-improvement mechanism,” Bavaria said. Its steering committee is now using the TAVI data (as well as the transcatheter mitral valve repair data) to develop risk models and composite outcome measures that can provide the basis for public reporting of outcomes by institution.

Low-Risk Patients: Proceed With Caution

Also speaking during the EACTS plenary, Friedhelm Beyersdorf, MD (University of Freiburg, Germany), presented 5-year outcomes among the more than 18,000 patients enrolled in the GARY registry, focusing on a propensity-matched subset of 1,820 patients treated with TAVI and the same number treated surgically.

While in-hospital outcomes were no different between groups back in 2011/2012 (with the exception of PPM implantation), 5-year survival in the matched cohort was significantly better for patients treated surgically than it was for patients who underwent TAVI (HR 1.51;95% CI 1.35-1.68).

That trend is similar, said Beyersdorf, to that reported in the 2019 OBSERVANT registry in low- and intermediate-risk patients, and to French data on high-risk patients from 2018. But it’s in stark contrast to 5-year results from the randomized trials such as PARTNER 1 in high-risk patients, the CoreValve pivotal high-risk trial, and the PARTNER 2A trial.

Speaking with TCTMD, Beyersdorf stressed the big differences between registry patients and randomized trials, with the former including the many types of patients excluded from the latter. His best guess is that the actual mortality risk difference between TAVI and SAVR is “definitely somewhere in between.”

Discussing the GARY data with TCTMD, Bavaria stressed that it’s difficult to apply the findings to contemporary patients, because the devices were first-generation and operator experience was in its infancy. “Those patients were pretty early in the TAVR experience, and I think that's the major limitation of that database at this point,” he said. “The data is real, but it's just a little old.”

Prendergast agreed, telling TCTMD that TAVI was very new to centers a decade ago—CT planning and sizing of devices was not universal, many centers were early in their learning curves, vascular complications were much more common, and first-generation devices had bigger problems with paravalvular leak—all of which he said “portends worse outcomes.” In addition, a full 25% of procedures back then were done using the transapical approach, which has been all but abandoned in most centers.

“Contemporary low-risk practice is very different from a retrospective determination from the GARY registry,” Prendergast summarized. That said, he continued, “I think it's important that the surgeons continue to remind the interventional cardiologists of the importance of long-term durability. . . . We all know that deep down, but interventional cardiologists have a shorter attention span in terms of new technologies and new procedures.

“We really do need the 5, the 8, and the 10-year long-term outcome data from both registries and randomized controlled trials,” he continued, “so that we can have utmost confidence when we speak to our younger low-risk patients about the best option.”

Beyersdorf, who was a chair of the recently released European Society of Cardiology/EACTS heart valve guidelines, urged both surgeons and cardiologists to follow the recommendations. “We really fought like crazy for every word in these guidelines, and at the very end, we came up with a result which is now published, and one should adhere to these recommendations,” he said. The data they are based on is “a mixture between registry studies and randomized trials. It's not perfect, nothing's perfect, but it's the best we have right now.”

Volume Trends

As the field awaits longer-term data to inform decisions in younger, lower-risk patients, the more contemporary numbers presented by Bavaria show a clear jump following the US Food and Drug Administration decision (and subsequent coverage) of low-risk procedures. TAVI implants rose from 59,420 in 2018 to 73,773 in 2019 and 77,348 in 2020. That figure increases to 82,075 when the growing number of valve-in-valve procedures are included in the tally.

Those rising TAVR numbers are having an impact on open-heart surgical procedures, said Bavaria, particularly after the approval of low-risk TAVI, with all surgical aortic procedures seeing a “substantial decline” relative to TAVI after 2018.

Median age is also steadily decreasing among the low-risk TAVI group, from 78 back in 2014/2015 to 76 in 2019 and 75 in 2020. STS score has also dipped from 3 to 2 over this same period. “This is going to be a very important number in comparing TAVR and SAVR outcomes as we move forward,” he said.

And while low-risk TAVI procedures made up less than 2% of procedures before 2018, expanded indications led to a marked increase such that low-risk patients now make up 28% of the overall cohort of patients heading for TAVI.

Asked whether that represents an appropriate shift, Bavaria said that depends on the perspective you take. On the one hand, the jump makes clear that prior to the expanded indications, US operators were largely “following the law” and restricting TAVI access to intermediate- and high-risk patients. “So it that an appropriate jump based on FDA approval? Yes. Is it an appropriate jump based on the guidelines that just came out, both the European and American? I'm not so sure,” he commented.

Some of these low-risk procedures “may be significantly off-guideline, but it's hard to tell,” Bavaria conceded.

The third way to answer the question of appropriateness is even less clear, he continued. “Do we really have enough long-term data to justify a jump of that magnitude in the low-risk patients? That we don’t know. . . . It's something that we in the TAVR community are going to have to figure out over the next few years, as the 5-year low-risk trials and other data sets come in, is whether we're doing the right thing or not.” 

One thing that is already clear as a bell is the impact of the COVID-19 pandemic. “As of June 2021, only 36,000 procedures were captured in the registry,” Bavaria noted, while surgical AVR procedures dropped by nearly 20% between 2019 and 2020.

This decline “points to a really a serious ethical issue that our country and other countries are going to have to confront, and that is this idea that a COVID death is higher priority than a cardiovascular life,” Bavaria told TCTMD, calling the shortfall a “serious byproduct” of COVID-19 policies.

Not all of the excess mortality reported in the US and elsewhere for 2020 is due to COVID-19, and at least some proportion of this will be due to people dying of severe untreated aortic stenosis, he said. “Critical aortic stenosis is a lethal, lethal disease and not to have these decreased number of procedures done likely means that those that are not getting them are dying.