TAVI Explantation Remains a Risky Proposition, Two New Studies Show

Two new studies are emphasizing the importance of appropriately choosing between TAVI and surgical aortic valve replacement in patients who have symptomatic severe aortic stenosis, with both highlighting the risks of taking out the transcatheter heart valve (THV) if it fails.

In the first study, an analysis of 396 patients included in the EXPLANTORREDO-TAVR International Registry, TAVI explantation was associated with a significantly higher risk of death at 30 days and 1 year when compared with having a second THV implanted. For the second study, which was an analysis of patients undergoing SAVR included in the Society of Thoracic Surgeons (STS) Database, surgical patients who’d initially been treated with TAVI had a significantly higher risk of dying during the procedure than those undergoing their second SAVR.

The papers, both published this week in JACC: Cardiovascular Interventions, emphasize the importance of “lifetime management,” with experts suggesting that if patients are expected to outlive an index THV, and their anatomy is unsuitable for redo TAVI, they should undergo SAVR first to treat aortic stenosis.

“With younger patients undergoing TAVR, we have to start thinking more seriously about, number one, what is the ideal first valve intervention in these younger patients, and number two, when the valve fails, what are the options that can be done safely and be potentially durable,” Gilbert Tang, MD, MSc, MBA (Mount Sinai Health System, New York, NY), lead author of the EXPLANTORREDO-TAVR analysis, told TCTMD. “Right now, the TAVR explant is a high-risk operation, it’s a more complex operation, and you can see that from the data.”

Right now, the TAVR explant is a high-risk operation, it’s a more complex operation, and you can see that from the data. Gilbert Tang

Shinichi Fukuhara, MD (University of Michigan, Ann Arbor), senior author of the STS analysis, said it’s becoming increasingly more common for people to “outlive the longevity of their first bioprosthesis,” but that there are no guarantees that redo TAVI can be performed after the first procedure.

Their data, he told TCTMD, highlight the risks of performing surgery after an initial TAVI. “Redo SAVR after SAVR is not as risky,” said Fukuhara. “After a high-quality SAVR, the feasibility of a valve-in-valve with TAVR is very, very high. There are a few good reasons why SAVR as a first approach is more favorable for younger people.”  

High-quality SAVR, he explained, would include aggressive enlargement of the aortic root to optimize a future valve-in-valve procedure with TAVI.

Tsuyoshi Kaneko, MD (Washington University School of Medicine, St. Louis, MO), one of the investigators for the EXPLANTORREDO-TAVR registry, said TAVI explantation has consistently been associated with mortality rates ranging from the mid-teens to over 20%, which “has raised some eyebrows.” However, it remains challenging to understand what it actually means.

“The easy message it sends is that if you need to have TAVI explanted, then the risk is going to be high, so maybe for younger patients we shouldn’t be doing TAVI,” Kaneko told TCTMD. “I think the equation might be a little more complex than that. The TAVI-explant patients tend to be older, even though the age is in the mid-70s in both studies. They are not a young population. Oftentimes, the explant time tends to be shorter, meaning they had earlier degeneration or some sort of complication from the original procedure, like large paravalvular leak. So, are we comparing apples to oranges, either in a repeat SAVR group or a redo-TAVR group? I think that’s the big question.” 

Higher Risks With Explantation

For the EXPLANTORREDO-TAVR registry, investigators wanted to look at the outcomes of TAVI patients undergoing explantation versus those receiving a second THV. A preliminary snapshot from this registry was presented at TVT 2022.

Overall, the incidence of THV failure among 66,760 patients treated with TAVI was 0.59%, a percentage that increased year over year during the 2009-to-2021 study period. Of the 396 THV failures, 215 patients underwent redo TAVI and 181 were treated surgically with THV explantation. The median time from the index TAVI to valve failure was 31.5 months, but the TAVI-explant group had a shorter median time from the index procedure to reintervention (17.6 months vs 45.7 months with redo TAVI; P < 0.001).  

The indications for reintervention were structural valve degeneration, paravalvular leak (PVL), severe prothesis-patient mismatch, delayed THV migration, and THV thrombosis. Structural valve degeneration was the most common reason for reintervention in patients undergoing redo TAVI (63.7% vs 51.9%; P = 0.023), while severe prosthesis-patient mismatch was more likely in the TAVI-explant group (17.1% vs 0.5%; P < 0.001). There was a similar incidence of moderate or worse PVL between groups. Balloon-expandable and non-balloon-expandable valves accounted for 40.2% and 59.8% of the reinterventions.

Overall, intraoperative and in-hospital mortality was 0.5% and 6.8%, respectively. In-hospital mortality was 11.6% with TAVI explantation versus 2.8% with redo TAVI (P = 0.001). At 30 days, the mortality rates in the TAVI-explant and redo-TAVI groups were 13.6% and 3.4%, respectively (P < 0.001). At 1 year, the respective mortality rates were 32.4% and 15.4% (P = 0.001). Stroke rates, however, were similar between the two reinterventions at 30 days and 1 year. In a landmark analysis from 30 days to 1 year, mortality rates with the two reinterventions were not significantly different.

“If you survive the initial operation for TAVR explant, your prognosis is actually similar to what it would be if you performed redo TAVR,” said Tang. “If these patients could be referred earlier when they’re less sick, their chance of success, their outcomes would likely improve. What we’re seeing is that they’re being referred a little bit too late because they’re considered too risky for surgery, but they don’t really have any other options, so by the time they get to surgery they have really advanced conditions.”

Tang noted there was a relatively higher rate of failure with the newer-generation valves, such as SAPIEN 3 (Edwards Lifesciences) and Evolut R/PRO/PRO+ (Medtronic), but said that’s only because more of these are being implanted in younger patients. With previous iterations, patients who received them were older and sicker and didn’t survive long enough to need a reintervention.

“Interestingly, the second valve put in [with redo TAVR] tends to be the opposite valve,” said Tang. “If they put in a balloon-expandable valve first, more people will put in a self-expandable valve next. It’s about a 60-40 split, though, so it’s not completely one versus the other. Either platform appears to work with whatever first valve failure.”

STS Database Analysis

In the second study, which was first presented at TCT 2022, Robert Hawkins, MD, MSc (University of Michigan), and colleagues turned to the STS Database to define the comparative risk of SAVR after prior TAVI or SAVR.

The complete cohort has been expanded from the initial presentation and now includes 31,106 patients undergoing SAVR. Of these, 29,306 had a redo SAVR (SAVR-SAVR) and 1,126 had SAVR after prior TAVI (TAVI-SAVR). The cohort also included 674 patients who required SAVR after a prior a valve-in-valve procedure (SAVR-TAVI-SAVR). In general, the TAVI-SAVR and SAVR-TAVI-SAVR patients were older, had a greater comorbid disease burden, had more multivalve disease, and had more frequent coronary artery disease.

In the unadjusted analysis, operative mortality was 17% for patients undergoing SAVR after failed TAVI. In contrast, operative mortality was 12% in those undergoing surgery after a prior valve-in-valve and 9% in those undergoing a second SAVR. In an adjusted model, those undergoing SAVR after an index TAVI had a 53% higher risk of operative mortality compared with the SAVR-SAVR patients (OR 1.53; 95% CI 1.14-2.06). In contrast, the SAVR-TAVI-SAVR group did not have a higher risk of mortality when compared with the SAVR-SAVR group.

When researchers excluded the SAVR-TAVI-SAVR group to limit heterogeneity and baseline risk differences, the operative mortality risk was three times higher in the TAVI-SAVR group than in those who had a second SAVR (15% vs 5%, P < 0.0001). In the risk-adjusted analysis, operative mortality was twofold higher in the TAVI-SAVR group. These results were also observed in a propensity-matched analysis of 433 patient pairs, with mortality 11.3% in the TAVI-SAVR group vs 6.7% in the SAVR-SAVR patients. In the propensity-matched analysis, the was no difference in the rate of unplanned aortic, mitral, or CABG surgery between groups.

To TCTMD, Fukuhara highlighted the number of patients with a valve-in-valve procedure who required reoperation. Data from the STS/ACC TVT Registry suggest that approximately 15,500 patients in the US underwent a valve-in-valve procedure (TAVI in SAVR) between 2013 and 2019, with 4,500 treated in the latest year. Based on this denominator, 674 patients undergoing SAVR-TAVI-SAVR at their center is quite high, said Fukuhara. 

Although he stressed these are assumptions based on available data, Fukuhara said the “high-volume of failed valve-in-valves requiring reoperations is a red flag.” He stressed that valve-in-valve failure is dictated by the initial SAVR, emphasizing the importance of surgeons considering future procedures. “If we put a small bioprosthesis during SAVR, valve-in-valve TAVR gets difficult,” said Fukuhara. “It’s not a great option for these people.”  

Making a Smart First Choice

With TAVI now clinically indicated across the risk spectrum, younger and younger patients are being treated. Knowing that they will outlive their first device makes discussions with the heart team critical. Interventional cardiologists and surgeons need to take a hard look at the patient’s characteristics to determine what should be the first strategy for symptomatic severe aortic stenosis, say experts.

In an editorial, Giuseppe Tarantini, MD, PhD, and Tommaso Fabris, MD (both University of Padua Medical School, Italy), say the best candidates for SAVR are likely those patients expected to outlive their THV and “those with the highest risk for coronary obstruction and sinus sequestration secondary to low coronary takeoff and small aortic root dimensions.” New models to help the heart team more accurately predict life expectancy are necessary, they add.

Kaneko said there are several limitations to the new data, highlighting the potential for selection bias. Patients with structural valve deterioration without PVL, and who have anatomy suitable for a second THV, will almost always undergo redo TAVI. He pointed out that many TAVI-explant patients require concomitant procedures, placing them at higher risk of morbidity and mortality.

Right now, said Kaneko, “it’s a little shortsighted to say that we shouldn’t be implanting TAVRs in a younger patient population.” He noted that TAVI technology continues to advance, with leaflet modification and leaflet resection techniques, that will facilitate more redo TAVIs in patients not currently candidates for a second THV.

Repeatability is really important. Shinichi Fukuhara

Fukuhara said a lot of research to date has focused on patients undergoing redo TAVI, noting there are limited data on the outcomes of patients undergoing explantation. With the inherent risks of explantation, he stressed the importance of thinking beyond the first valve.

“If TAVR is the first approach, and it happens quite often nowadays because patients say they don’t want open-heart surgery, that first TAVR needs to be oriented for the second TAVR,” said Fukuhara. “Repeatability is really important. If we feel that there is no way that this patient can have redo TAVR after the first TAVR, which can be obvious in patients with a small aortic root, then we really don’t recommend TAVR [as a first procedure]. That’s what we’re doing at the University of Michigan.”

Tang had similar thoughts. If the CT scan suggests operators can achieve a good first TAVI, they must then turn their attention to whether a second TAVI can be performed down the road based on the patient’s anatomy. “If the answer is yes, then TAVR is a reasonable first step because you know you can do a redo TAVR in the future with either platform, he said.

However, if the anatomy will preclude a redo-TAVI procedure, “then it’s an important discussion with the heart team and patient,” said Tang. “They need to know that they’re not going to avoid open-heart surgery by getting TAVR today.” TAVI is not an unreasonable first procedure if the patient is willing to take on surgery in the future, whether it’s 8 or 10 years, depending on when that first TAVI fails, and interventionalists can achieve an excellent result with the first implant.

“Of course, you can also tell the patient that if they have surgery now, they have at least 10 to 15 years of durability, and then they can get TAVI later with a valve-in-valve,” said Tang. “That’s also not unreasonable.”