Device Explants After TAVR Rare but Risky

The number of transcatheter heart valves explanted after TAVR in the United States is encouragingly low, but surgical explantation remains a high-risk procedure associated with a significant risk of death at 30 days, according to a large analysis from a Medicare administration database.

Among 132,515 patients treated with isolated TAVR between 2012 and 2017, surgeons removed just 227 (0.2%) valves, with bioprosthetic valve failure (79.3%) and endocarditis (20.3%) being the two reasons provided for explanting the transcatheter valves.

“The low rate is very reassuring, but I do think we need to take that with a grain of salt,” lead investigator Tsuyoshi Kaneko, MD (Brigham and Women’s Hospital, Boston, MA), told TCTMD. “These patients were at extreme or high risk for surgery, at least up until 2015, so there may be some cases that were turned down by surgeons even if they were needed because the patients were too high risk. However, I do think the percentage is low enough, at least in this short-term follow-up, that it is reassuring.”  

When surgery is needed, however, it is associated with substantial risks, said Kaneko. In their analysis of the Medicare Provider Analysis and Review database, which Kaneko presented earlier this week as a late-breaking clinical trial during TVT Connect, the 30-day mortality for patients who had their transcatheter valve explanted was 13.2%. Additionally, 5.7% of patients had a permanent stroke, 29.1% had acute kidney injury, and 11.5% developed complete heart block. In-hospital bleeding complications were observed in 55.9% of patients, and the mean length of stay in the ICU was 5 days. Mortality at 1 year was 22.9%.

In a multivariable, time-dependent Cox regression analysis, surgical explantation was associated with a more than fourfold higher risk of death compared with patients who didn’t have their valve removed (HR 4.03; 95% CI 1.81-8.98).

“These were very, very high-risk procedures,” said Kaneko. “Some of that risk comes with the patient characteristics…Then you go in, and TAVR valve comes out relatively easy if it’s early enough, but the longer it’s in, or with certain types of valves, it can be more difficult to take out. In some of cases, the valve may be adherent to the aorta and require some complex aortic [procedure]. In some instances, it might be attached to the mitral valve leaflets, which would require mitral valve surgery. Some of the techniques are not really well understood amongst the surgeons yet. There is a lot more to learn.”

In the series, the median time to surgical explant was 212 days. Breaking that down further, 70.9% of explants took place in the first year and 46.3% within 6 months of the index TAVR. Just 8.8% of valves were removed within 30 days.     

The researchers performed a number of sensitivity analyses, including a look at the adjusted survival in the explant cohort based on the timing of explantation and TAVR era. There was no significant difference in survival among patients who had their valve explanted within 6 months of the index TAVR versus those who had the valve removed beyond that time. Similarly, when looking at the TAVR era, which allowed them to explore the risks in patients without a high/extreme risk of surgery at baseline, they saw no difference in cumulative survival among patients with explanted valves initially treated between 2012-2014 and 2015-2017. Finally, there was no difference in mortality among explanted patients with a high or low burden of baseline comorbidities.

Just a couple of weeks ago, data from the Society of Thoracic Surgeons adult cardiac surgery database showed a similarly high rate of mortality among patients who required surgical aortic valve replacement following TAVR. In that series, the operative mortality rate—death within 30 days of the procedure or before discharge—was 17.1%. Unlike the Medicare analysis, the transcatheter valves were removed a median of 2.5 months after the index TAVR, with the most common reasons being paravalvular leak, structural prosthetic deterioration, failed repair, sizing/position issues, and endocarditis.

Difficult to Tease Out Reasons for Explantation 

To TCTMD, Kaneko said that given the absence of detailed data on the Medicare patients, they are unable to tease out the exact reasons for surgical explantation beyond a broad definition of bioprosthetic valve failure. “About 20% were endocarditis, which is the population in whom I think will require more surgical explantation,” he said. “The infected valve has to come out. For the rest, we don’t know if it was structural valve deterioration, if it was paravalvular leak, or if some of the patients required some other valve surgery.”

For instance, with mitral valve surgery, surgeons will sometimes remove the transcatheter valve to obtain access and get better visualization. “We have to look at that 80% with a little bit of caution,” said Kaneko. “We don’t know if all 80% had valve stenosis or valve failure. But we all know that the first-generation valves did have higher rates of paravalvular leak, and we do operate for those reasons.” 

The researchers also were unable to look at the rates of explantation by valve type. During the discussion following his presentation, Kaneko said there are reports that the removal of CoreValve (Medtronic), given that it extends through the sinus and ascending aorta, does require a larger operation.

Julinda Mehilli, MD (Ludwig-Maximilians University, Munich, Germany), one of the session’s discussants, said the primary reason for surgical explantation of TAVR devices is recalcitrant restenosis of the valve. At her center, patients with paravalvular leak are typically referred for a valve-in-valve procedure. They have had relatively few cases of endocarditis over the past decade, she said. 

Overall, Kaneko said the Medicare data are just a start to understanding the incidence and risks of surgical explantation. Nonetheless, the mortality finding at 30 days is a reliable measure, he added, and the rate “should warn us a little bit” about implanting valves in younger patients.

“I don’t know if younger patients will have a higher mortality like in this series, but it clearly shows that this isn’t a low-risk procedure,” he said. “If we have to go back in for a TAVR explant in a patient that was implanted at age 65, and they’re now 75, what are those risks going to be if they can’t get a TAV-in-TAV procedure? We have to watch this carefully because I think it will have an effect on the lifetime management of these younger, low-risk patients.”