High Mortality but Durable Results With Valve-in-Valve TAVR

SAN DIEGO, CA—One-third of patients who undergo valve-in-valve TAVR are dead at 3 years, but patients who survive have significant and durable improvements in valve hemodynamics, functional status, and quality of life, according to new data from the PARTNER II aortic valve-in-valve registry.

Presenting the results at TCT 2018, lead investigator John Webb, MD (St. Paul’s Hospital, Vancouver, Canada), said that valve-in-valve TAVR is a “viable alternative” for patients with failing bioprosthetic valves not eligible for reoperation. From a technical standpoint, not only are these procedures “somewhat easier” than implanting a native transcatheter aortic valve, Webb said, there also are “dramatic” clinical benefits of the valve-in-valve TAVR procedure for degenerated surgical bioprostheses.

The late follow-up included data from the nested registry and continued access registry from PARTNER II, totaling 365 patients who underwent for valve-in-valve treatment with a 23- or 26-mm Sapien XT transcatheter heart valve (Edwards Lifesciences). At 3 years, the primary composite endpoint of all-cause mortality or stroke was 35.6%. Mortality increased from 12.1% at 1 year to 32.7% at 3 years. That the stroke rate did not accrue over time suggests there is no “late price” with the procedure, said Webb.

“I would point out that these are patients with multiple comorbidities,” said Webb, referring to the high rate of mortality.

The patients included in the valve-in-valve registry had an initial bioprosthetic valve less than 21 mm as well as greater than mild paravalvular regurgitation, left ventricular ejection fraction less than 20%, or an estimated life expectancy of less than 2 years. The average age of patients was 79 years, the mean STS score was 9.1%, and more than 90% of patients had NYHA class III/IV heart failure symptoms.

The majority of patients were treated with the 23-mm Sapien XT valve. As a result, the effective orifice area (EOA) is small at 1.2 cm2, but the results were durable to 3 years, said Webb. Similarly, the mean pressure gradient was 17.7 mm Hg at discharge/30 days after the valve-in-valve procedure, but this reduction was maintained to 3 years. The proportion of patients with total and paravalvular aortic regurgitation (≥ moderate) was dramatically decreased from baseline to the discharge/30-day assessment and this reduction was also maintained to 3 years.

In late follow-up, the change in function and quality of life was also durable, said Webb. Three years after the valve-in-valve procedure, 86% of patients had NYHA class I/II heart failure symptoms and the improvement in the Kansas City Cardiomyopathy Questionnaire (KCCQ) quality-of-life score was not significantly different from the summary score assessed at 30 days.

Chandan Devireddy, MD (Emory University, Atlanta, GA), who was not involved in the study, said valve-in-valve procedures require “some know-how” about the unique properties of the surgical valves and the technical challenges they might present when implanting a transcatheter valve. “They’re all a bit different,” said Devireddy. “The presence of the surgical valve also assists the operator in setting a target for deployment of the TAVR valve. Other than that, the workup is generally similar.”

Patients who require a valve-in-valve procedure tend to be quite sick, he added, which is borne out by the high rate of mortality at 3 years.

“However, the recovery is profound,” said Devireddy. “It’s a life-changing event for these patients if you can do this. . . . I think the same efficiencies and rapid improvement that we’re seeing with standard TAVR can be seen with valve-in-valve procedures as well. In some ways, they have the most to gain, especially if you consider the alternatives, which is nothing at all, or a horribly morbid redo sternotomy and valve replacement that would likely involve extensive hospital stay and prolonged rehabilitation.”

‘Some of the Easiest Procedures to Do’

In their regional program, which operates as a “hub and spoke” model, Webb said valve-in-valve TAVR was initially restricted to just one institution. However, they have since abandoned that protocol to allow the four “spoke” TAVR-capable hospitals to perform the valve-in-valve procedure. “We had to give that up because these are some of the easiest TAVR procedures we do,” Webb told TCTMD. “We have a radiopaque valve, we know the angle and the size, and it seals well. We don’t get annular rupture and you don’t need pacemakers very often.”

Additionally, an analysis from the TVT Registry has suggested that clinical outcomes were better with valve-in-valve implants compared with native implantations. “I think there’s a knowledge base required that routine TAVR operators might not have, but it can be taught, it can be learned,” said Webb. “It’s not a difficult procedure when you know how to do it.”

Jeffrey Popma, MD (Beth Israel Deaconess Medical Center, Boston, MA), questioned whether physicians should be “reconsidering” anticoagulant therapy after valve-in-valve procedures. “One of the things we’ve learned from the leaflet thrombosis data is that one of the risk factors is valve-in-valve procedures,” said Popma. “Clinically, we do see cases where thrombus does form in the nidus with all the material that’s there.”

In response, Webb said they observed no signs of leaflet thrombosis playing a role in clinical reinterventions. That being said, the researchers weren’t looking for evidence of leaflet thrombosis with dedicated transesophageal echocardiography or CT scans. “Personally, I suspect that we should be routinely anticoagulating all of our valve-in-valve implants.” Webb said there is no guidance on the use of anticoagulation after a valve-in-valve procedure, nor was it recommended in PARTNER 2, but some physicians are using it following valve-in-valve TAVR.

Devireddy said the use of anticoagulation varies by operator and there’s no dedicated strategy known to be advantageous in these patients. “We do think there likely is a higher tendency for thrombosis, given whatever the fluid dynamics are for valve-in-valve, but does mean we should use prolonged anticoagulation? Anticoagulation for a month? Or stick with dual antiplatelet therapy?” he said. “Hopefully we’ll know more in the future but right now most operators are practicing off the cuff.”   

To TCTMD, Webb said the trial was performed before the concept of “fracturing” the preexisting surgical valve to help with implantation and that this was not performed in these valve-in-valve patients. “If you have a surgical valve 21 mm in diameter, then the internal diameter might only be 18 mm in diameter,” he explained. “Now you’re putting a transcatheter valve inside a very small orifice that can’t expand. There had been concerns about increased pressure gradients across the valve or that the valve might be less durable because the leaflets can’t open and close as they were designed.”

Still, the results suggest that while there might be residual increases in the aortic pressure gradient, the clinical benefit was dramatic and there were no signs of valve failure, said Webb.