Good Outcomes With Transcatheter Mitral Valve-in-Valve, With Caveats

Experts have some concerns about the now-published data, including what elevated mitral gradients might mean long-term.

Good Outcomes With Transcatheter Mitral Valve-in-Valve, With Caveats

For patients with degenerated bioprosthetic valves, transcatheter mitral valve-in-valve (MViV) using Sapien 3 (Edwards Lifesciences) provides low rates of mortality and improves heart failure symptoms, now-published data from the Society of Thoracic Surgeons/American College of Cardiology TVT Registry show.

Moreover, use of a transseptal versus transapical approach was associated with lower mortality at 1 year (HR 0.58; 95% CI 0.37-0.90), according to lead author Brian Whisenant, MD (Intermountain Heart Institute, Salt Lake City, UT), and colleagues.

The findings were published online July 29, 2020, ahead of print in JAMA Cardiology. A similar analysis with one additional month added to the study period was presented by Mayra Guerrero, MD (Mayo Clinic, Rochester, MN), at TCT 2019.

“When bioprosthetic surgical mitral valves fail, transseptal MViV can be performed with low risks, rapid recovery, and excellent in-hospital, 30-day, and 1-year outcomes,” Whisenant told TCTMD in an email. He noted that the observed rate of 30-day all-cause mortality in patients undergoing a transseptal procedure was 5.0%, half of what would be expected based on the mean STS score at baseline (11.0%).

“Transcatheter mitral valve-in-valve replacement is currently indicated for patients with a high risk for surgical reoperation,” he continued. “However, our results support transcatheter mitral valve-in-valve replacement as an option for most patients with failed surgical valves, independent of the patient's surgical risk. We anticipate that this data will support an expanded indication and much broader use of this less-invasive option.”

Speaking with TCTMD, Anita Asgar, MD (Montreal Heart Institute, Canada), who wrote an accompanying commentary, said this is the largest collection of data available on MViV and is “a great first step [in] trying to understand the place of the technology.”

Safety was acceptable for this population of high-risk patients, but there are some concerns, she said, pointing primarily to the hemodynamic result of the procedure. “The disappointing thing from my point of view is once patients had this procedure, their mean mitral gradient was around 7 mm Hg,” down from a mean of 12.7 mm Hg at baseline, Asgar said. “So an improvement, but far from what I would consider this massive procedural success because you are leaving patients with a significant gradient despite this intervention.”

That’s been seen in smaller MViV studies, as well as with aortic valve-in-valve procedures, she noted, “but obviously we know that high residual gradients on the mitral side is not the same thing as on the aortic side. It could be of potentially more concern.” That concern has some support from the current study, in which patients who received smaller valves and had higher residual gradients had increased mortality at 1 year. “These are early signals of things we need to be prepared for and things we need to understand if we’re going to propose this procedure to a patient,” Asgar said.

Transseptal Outcomes Better

For the study, the investigators used data from the TVT Registry and the US Centers for Medicare & Medicaid Services. The analysis included 1,529 patients (mean age 73.3; 59.1% women) who, between June 2015 and July 2019, underwent transseptal or transapical MViV (86.7% transseptal) with the Sapien 3 valve at 295 US hospitals.

The primary safety endpoint was procedural technical success, defined by the Mitral Valve Academic Research Consortium (MVARC) criteria upon exit from the hybrid suite—patient is alive with successful access, delivery, and retrieval of the delivery system; successful deployment and correct position of the first intended device; and freedom from emergency surgery or reintervention associated with the device or access procedure. This was achieved in 96.8% of cases. Procedural complications were infrequent and included cardiac perforation (1.1%), left ventricular outflow tract (LVOT) obstruction (0.9%), stroke (0.7%), and device embolization (0.3%).

We know that high residual gradients on the mitral side is not the same thing as on the aortic side. It could be of potentially more concern. Anita Asgar

Transseptal versus transapical access was associated with a shorter median length of stay (2 vs 6 days) and a greater likelihood of being discharged to home (82.5% vs 59.1%; P < 0.001 for both).

Overall, the rate of all-cause mortality was 4.0% in the hospital, 5.4% at 30 days, and 16.7% at 1 year (the primary efficacy outcome). Of note, 30-day mortality was lower than expected based on STS score for both transseptal and transapical procedures. Mortality outcomes tended to be better with transseptal access at each time point, becoming a significant advantage at 1 year (15.8% vs 21.7%; P =0.03).

The procedure was associated with a marked drop in the percentage of patients with NYHA class III/IV heart failure (from 87.1% at baseline to 9.7% at 1 year) and improvement in quality of life as defined by the Kansas City Cardiomyopathy Questionnaire (39.4-point improvement by 1 year).

Whisenant pointed out that “these results were achieved with a median implant volume of only three implants per enrolling site, with steadily improving outcomes over the time of this analysis.”

Unresolved Issues

Pointing to the in-hospital mortality rate of 4.0% overall, Danny Dvir, MD (University of Washington, Seattle), who was not involved in the study, said outcomes were “quite reassuring” in this cohort of high-risk patients. He noted that presentations 5 to 10 years ago had in-hospital mortality rates approaching 10%.

The study also shows that the transseptal versus transapical approach yields better outcomes, he said.

But, like Asgar, Dvir had concerns about the mean gradient of around 7 mm Hg after the procedure, which he said “is not an optimal result” and indicates that the patient still has moderate-to-severe mitral stenosis. According to MVARC, he added, the mean gradient should be below 5 mm Hg. That needs to be considered when thinking about who should undergo this procedure, he said: “What would be the durability of procedures that finish on average with mean gradient of 7 mm Hg? What would be the rate of reintervention? What would be the functional class of patients in follow-up if they finished the procedure with moderate-to-severe mitral stenosis?”

Operators need to think about how to optimize the procedure to bring the gradient down, Dvir said, considering factors like implantation position and depth, valve size, and use of bioprosthetic valve fracture in the mitral position to allow for better expansion of the transcatheter heart valve. In addition, he said, it’s possible that anticoagulation should be used in all patients to improve postprocedural hemodynamics; in this study, 80.8% of patients were discharged on oral anticoagulants.

Asgar noted that the current study cannot answer questions about the long-term implications of having a postprocedural gradient of around 7 mm Hg. “Are we putting patients at risk for early valve failure? Are we increasing the risks for thrombosis of the valve or this hyperattenuation that’s been well described on the aortic side but to date we haven’t really started looking for it yet on the mitral side?” she said. “Given the mortality they’ve seen, it’s unlikely it’s going to be a good thing, but the reality is we just don’t know.”

Responding to the concerns about the postprocedural gradient, Whisenant pointed to the gains seen in this cohort in terms of NYHA functional class and quality of life. “Our data, as well as my personal experience with these patients, is that the mildly increased gradients noted after MViV are not associated with heart failure symptoms or other adverse events,” he said.

MViV or Redo Surgery?

Asked about the proportion of patients with degenerated bioprosthetic valves who should undergo MViV versus repeat surgery, Dvir noted that the majority of patients in this situation are relatively old. Patients over age 65 are already at high risk for repeat surgery, so MViV would be a good option. But in a 45-year-old patient without significant comorbidities, Dvir said he’d still consider a redo operation because of the unknown durability of MViV. “With some patients, I would still consider having repeat mitral valve surgery and then think about the option of putting in a mechanical device,” he said.

If one were to opt for a MViV procedure, Dvir stressed, hemodynamic outcomes would have to be optimized, a point he has made before. “If we finish with moderate-to-severe mitral stenosis, I think that we do not do justice to patients,” he said.

As for which patients might be suitable for surgery, Whisenant said that “some younger patients with failed bioprosthetic mitral valves may opt for surgery with a mechanical valve as a more definitive solution. Rare patients have unfavorable anatomy for mitral valve-in-valve, such as a high risk of LVOT obstruction, and may consider surgery depending on their surgical risk and the operator’s comfort with alcohol septal ablation and/or LAMPOON.”

Asgar noted that if there is a patient who is at high surgical risk or has been refused for surgery, “your options are limited, and I think it’s important to advise patients that we do have a treatment that’s potentially available for them, and that is transcatheter mitral valve-in-valve as opposed to medical treatment, for which the outcomes will likely be poor.” But patients need to be adequately informed about what is known and MViV at this point and about the expected procedural results.

Many patients with degenerated bioprosthetic valves also have tricuspid regurgitation, Asgar noted. In this study, 55.7% had moderate-to-severe tricuspid regurgitation. There are still no good answers about what to do for this type of patient, she said.

Regarding MViV, Asgar said: “The onus is on us as clinicians and clinician-scientists to be able to generate the evidence that will allow us to say this is for everybody or perhaps more importantly to really discern for whom this technology is going to be the adequate treatment and for whom we may need to consider other options.”

  • Statistical analyses were performed by Edwards Lifesciences.
  • Whisenant reports personal fees from Edwards Lifesciences outside the submitted work.
  • Asgar reports consulting fees from Abbott, Edwards Lifesciences, and Medtronic.
  • Dvir reports being a consultant to Edwards Lifesciences.

We Recommend