Valve-in-Valve TAVI Bests Redo SAVR in Short Term

Valve-in-valve (ViV) transcatheter procedures for a failed aortic bioprosthesis are associated with a lower risk of in-hospital mortality compared with a repeat surgical aortic valve replacement, but that benefit is offset slightly by a higher risk of hospital readmission at 30 days and 6 months, according to the results of a new study of national US data.

Importantly, there was no difference in the risk of in-hospital stroke, postprocedural conduction disturbances requiring a pacemaker, major adverse cardiovascular events, or mortality between the two procedures during early follow-up to 6 months.

“Whenever we have patients in clinic, I try to talk about the three main risks associated with TAVR,” senior investigator Ankur Kalra, MD (Cleveland Clinic, OH), told TCTMD. “That’s the incidence of stroke, the incidence of pacemakers, and the incidence of vascular complications. If you look at the contemporary data from randomized clinical trials—not in bioprostheses but in native aortic valves—one could argue that the outcomes are better for TAVR compared with SAVR. The fact that we can extrapolate those findings now to valve-in-valve procedures, where one would assume patients have a heightened risk, is very promising.”

The retrospective analysis of the Nationwide Readmission Database (NRD) from 2016 to 2018, with first author Monil Majmundar, MD (Maimonides Medical Center, Brooklyn, NY), was published last week in EuroIntervention.

Tsuyoshi Kaneko, MD (Brigham and Women’s Hospital, Boston, MA), who along with Sameer Hirji, MD (Brigham and Women’s Hospital), has previously looked at NRD data to understand differences in clinical outcomes following redo surgery versus ViV TAVI, said the more-contemporary data contain at least one change from the earlier snapshot.

“There’s definitely a trend here that we’re seeing that we did not see before, which is the increase in valve-in-valve TAVR to where it’s surpassing the number of repeat surgeries,” Kaneko told TCTMD. Like their series, which used slightly older NRD data, early mortality was lower with ViV TAVI and complication rates were similar. In this new series, the in-hospital mortality rate was 1.2% with ViV TAVI versus 3.4% with repeat SAVR.  

“It goes along with what we’ve been seeing in other data,” said Kaneko. “Valve-in-valve TAVR is a very, very safe procedure, with very low short-term mortality. Repeat SAVR, on the other hand—we’ve reported data from the STS database, and there’ve been recent updates as well—the short-term mortality is 3% to 5%, and this new series goes along with that finding.”

Hirji agreed, saying that the new analysis provides reassurances for ViV TAVI, with some caveats. “We should take the data cautiously as we interpret these findings because we don’t know the long-term implications of these procedures,” he told TCTMD. “We want to make sure we’re thinking about the lifetime management of these patients.”    

More ViV TAVIs Than Repeat Surgeries

Current clinical guidelines recommend surgery for a failed aortic bioprosthesis (class I indication), but the Sapien XT balloon-expandable transcatheter heart valve (Edwards Lifesciences) and the CoreValve self-expanding platform (Medtronic) were both approved by the US Food and Drug Administration for ViV procedures in 2015 for patients at high or extreme risk for complications during surgery. 

In 2019, Kalra presented data from the STS database that showed repeat SAVR for failed bioprosthetic aortic valves was on the decline between 2012 and 2016 but dropped off precipitously after the FDA approval of ViV procedures. This newest analysis from the NRD, said Kalra, is a continuation of that research in that they wanted to compare early clinical outcomes of ViV TAVI versus repeat SAVR.

The researchers identified 6,769 patients who underwent either TAVI or SAVR for a failed aortic valve bioprosthesis. Of these, 3,724 patients were treated with the transcatheter approach and 3,045 with surgery. Those who underwent ViV TAVI were much older than those treated with surgery (mean age 79 vs 65 years; P < 0.001) and had a higher prevalence of hypertension, diabetes, hyperlipidemia, prior history of MI, stroke/TIA, ischemic cardiomyopathy, chronic heart failure, atrial fibrillation, prior PCI, prior CABG, pacemaker/defibrillator therapy, cancer, chronic pulmonary disease, and chronic renal failure.

“The valve-in-valve TAVI patients have a higher comorbidity burden and that’s why they’re poorer patients for repeat SAVR,” said Kalra. “That’s also why it’s so important to propensity-match this group because you want to compare apples to apples.”

In the propensity-matched model, the risk of in-hospital mortality was 58% lower in those treated with ViV TAVI compared with surgery (OR 0.42; 95% CI 0.20-0.90), but there was no difference in the risk of stroke or vascular complications. Cardiorespiratory complications and the risk of major bleeding were higher with surgery. In terms of the need for permanent pacemaker implantation, there was no significant difference between TAVI and SAVR.

The risk of hospital readmission was a relative 46% higher with TAVI at 30 days and 54% higher at 6 months, a significant difference compared with surgery.

“This is an important metric to report because it’s important for patients,” said Kalra. “Patients are inconvenienced if they are rehospitalized after a major procedure like getting another valve.” However, he pointed out that most of the readmissions were driven by major bleeding or vascular complications which were outside the index procedure or noncardiac infections, such pneumonia, bacteremia, and sepsis.

Kaneko said the rate of permanent pacemaker implantation with ViV TAVI has consistently been shown to be lower compared with native TAVI. Additionally, with repeat SAVR, removal of the bioprosthesis before the placement of the new valve can translate into higher rates of new permanent pacemakers compared with the initial surgery.

“It’s working in opposite directions and I think that’s what we’re seeing here with no difference between the two procedures,” said Kaneko. In terms of the higher rate of readmissions, Kaneko suspects propensity-matching was unable to capture the potential cofounders influencing the results.  

Future Studies

In clinical practice, Kalra uses the STS-Predicted Risk of Mortality (STS-PROM) risk score to determine whether a patient with a failed bioprosthetic valve would be an appropriate candidate for repeat SAVR. If the surgical risk is intermediate, high, or prohibitive, he’ll discuss the option of a ViV TAVI given the lower risk of up-front morbidity. He added that patients don’t want to undergo surgery again if they can avoid it.

“They don’t want another crack at their sternum, quite frankly,” said Kalra. “If it were me, would I want a repeat surgery? No.”

In terms of the trend toward physicians performing more ViV TAVI compared with SAVR in the database, Kaneko agreed with Kalra that patients are aware of the recovery process following a sternotomy and would prefer a minimally invasive ViV procedure over repeat SAVR. There might also be some hesitancy on the part of surgeons, noting that a second SAVR is not a straightforward procedure after previous cardiac surgery, said Kaneko.

“Also, there’s been a lot of recent papers showing that even in low-risk patients, these valve-in-valve TAVRs do quite well,” he said.

Hirji said that a variety of factors influence the treatment decision, but in general, older and frail patients tend to be better candidates for minimally invasive TAVI as opposed to SAVR. These new data provide much needed information for the heart team evaluating patients in need of a second valve, but Hirji stressed that the short-term outcomes need to be kept in perspective, particularly since there is limited medium- or long-term data. 

“For an older person, it might not matter but we don’t have much data on durability for valve-in-valve procedures,” he said. Additionally, he said the failed index procedure highlights the importance of selecting the appropriate initial treatment of symptomatic, severe aortic stenosis. 

To TCTMD, Kalra noted that they did not have data available on the primary valve types/sizing, anatomical characteristics of the replaced valve, type of TAVI valve used, or echocardiographic parameters, such as of aortic-valve area and gradients, among others. Also, he pointed out that the ViV patients may be at an increased risk of future valve thrombosis, so longer-term follow-up is needed, including study into the optimal antithrombotic/antiplatelet strategy. Additionally, the potential for future coronary access needs to be considered in these ViV cases and requires further study, he said.