TAVR For Failing TAVR Doesn’t Impress in Registry Data

Transcatheter aortic valve replacement (TAVR) for a degenerated TAVR valve appeared feasible and safe in the TRANSIT registry, although 5-year results from TAVR for failing surgical valves in PARTNER II looked “more optimistic.”

Both sets of data were reported at the virtual TCT Connect conference.

Data like these are “changing our algorithm for how we initially make decisions in our patients,” Rebecca Hahn, MD, of NewYork-Presbyterian/Columbia University Medical Center in New York City, told MedPage Today.

“The age cut off for putting in a valve has gradually moved down,” she added, “and in part that’s because…we know we can salvage those bioprosthetic valves with a transcatheter procedure that is relatively safe and has good outcomes, now we’ve got outcomes out to 5 years with a second-generation valve.”

As TAVR increasingly moves into younger, lower surgical-risk patients, long-term outcomes and options for repeat valve replacement are going to be important, said Luca Testa, MD, PhD, of San Raffaele Hospital in Milan.

TAVR After TAVR

In Testa’s TRANSIT data, a second TAVR bioprosthesis was implanted with 79% device success by VARC-2 criteria in 172 patients at a median of more than 3 years after the initial successful TAVR.

Most of the unsuccessful second TAVR was due to residual significant gradient in patients with severely stenotic degenerated valves, but no malpositioning of the second valve was seen, Testa reported.

He pointed to no coronary obstructions or aortic dissections and “extremely low” complication rates: one case each of annulus rupture and emergency surgery (0.6%) and three major procedural vascular complications total (1.7%).

At 1 year, the overall mortality rate was 10%, cardiovascular mortality 5.8%, stroke 3.5%, and 1.2% myocardial infarction.

Testa pointed to the 11% heart failure hospitalization rate and 12.8% in New York Heart Association (NYHA) class III/IV at 1 year as evidence of significant clinical benefit.

“We are talking about an elderly population where a first TAVR has failed, and we were in the position of deciding what to do next — trying a new TAVR without knowing, honestly, what was the outcome,” since there has not been data in such a setting, Testa said. “Medical therapy is not actually an option and surgery has been rejected years before.”

TRANSIT included 172 second TAVR cases out of the total of about 40,000 procedures done since 2008 at the 28 participating centers, largely in Europe but with four in North America. Mode of failure of the first bioprosthetic valve was stenosis in 57, regurgitation in 97, and mixed in 18.

Nearly all of the cases were done transfemoral, with predilatation in about one in five and postdilatation in 41%. All the patients had been at high surgical risk for their initial procedure. More than 80% were in NYHA class III/IV.

While further data is needed for lower risk and young patients (mean age was 79.9 in TRANSIT), Testa concluded that the kind of patients in his study could be “safely and successfully treated by means of a second TAVR.”

“However, I have to say I am not as optimistic about the results,” countered session co-chair Michael Borger, MD, PhD, of Leipzig Heart Center in Leipzig, Germany. “The STS score was 6, yet the 30-day mortality for this group was 7% and the procedural success rate of 79% is also lower than what we’ve seen in most other studies.”

TCT session discussant Vinayak Bapat, MD, also of NewYork-Presbyterian/Columbia, agreed. “I am less optimistic because of this,” he said.

These patients had a successful initial TAVR but then required another 2 to 4 years later, so what of their future another 2 to 4 years down the road?, he posited.

“Should we explant TAVRs and put surgical valves? Because these are intermediate risk patients, STS score of 6,” he noted. “Or should we do another TAVR that is probably going to last less than the previous TAVR lasted in these patients?”

Testa maintained that his data are “pretty reassuring” for elderly patients who outlive their TAVR. But for patients in their in 60s, for example, “of course, we need direct comparison with SAVR [surgical aortic valve replacement]. We don’t have enough data. We don’t know a bioprosthesis implanted surgically will last longer than a TAVR.”

Borger agreed that there may be something surgical risk scores don’t capture about a patient who has an initially successful TAVR that then fails within a few years. “There’s something about them that they’re not tolerating the valve well, and I’m not sure it’s the best strategy to go ahead and put a second valve in.”

Testa suggested that longer-term follow-up of the registry patients may help answer some of these questions.

TAVR after SAVR

In a separate presentation at the same session, 5-year PARTNER II trial data supported the long-term outcomes and durability of TAVR done for failing surgical bioprostheses, which were typically more than 10 years old.

Five-year mortality was similar between the high or extreme surgical risk valve-in-valve patients in PARTNER IIb and those with native aortic stenosis in the intermediate risk arm of PARTNER IIa (50.6% vs 45.9%, P=0.06), reported Hahn.

Among the 365 high and extreme surgical risk cases with a valve-in-valve procedure, hemodynamics appeared stable from years 1 through 5 for mean gradient and Doppler Velocity Index. Aortic regurgitation remained no more than trace in 86.9% and mild in 10.1% at 5 years.

TCT press conference moderator Ori Ben-Yehuda, MD, of the University of California San Diego, drew attention to the “astounding” improvement in Kansas City Cardiomyopathy Questionnaire scores that remained stable, with an average 74.2 on the 100-point scale at 5 years.

Structural valve deterioration resulted in adjudicated hemodynamic valve deterioration in 5.9% of patients and in valve fracture in 2.3%. Bioprosthetic valve fracture from any cause occurred in 4.7% of patients by year 5. These cases were nearly all in Sapien XT valves in the smaller 23 mm size that 69% of patients had gotten.

The only significant multivariate predictor of mortality risk, aside from higher STS score and male sex, was TAVR valve size (7% higher risk at 5 years with 23 vs 26 mm valves). However, Hahn cautioned that these patients differed in important ways, such as the 23-mm valves typically replacing smaller surgical valves and being done in a group with more prior cardiac events.

Still, it’s further evidence, Bapat said, that “we as surgeons, if we are putting in small valves we have to put in valves that will be expandable or do something else to facilitate the results.”

“These 5 year results are very encouraging,” commented Bapat. “The valve-in-valve results are more optimistic than TAVR-in-TAVR results.

The rate of bioprosthetic valve failure and structural deterioration are “really, really low” now with the newer generation of TAVR devices than tested in PARTNER II, and likely have even better outcomes than these implanted 5 or 6 years ago, Testa noted.

Borger cautioned also that “one big caveat” was that these are results from intermediate- and high-risk patients, whereas more data — possibly from a head-to-head trial — would be needed for low risk and younger bioprosthetic valve failure patients.

“In the low risk patient, we’re looking for much longer term outcomes than even 5 years,” Hahn said. She agreed that such information would help inform patients in selecting among their options for their first procedure.

And if TAVR-in-TAVR ends up panning out, she suggested, that also “may end up changing the early decision making process in those patients.”