SAVR After Failed TAVR: Rare and Risky, Registry Study Finds

It will be important to follow this data in more contemporary cases as TAVR moves to lower-risk patients, the lead author says.

SAVR After Failed TAVR: Rare and Risky, Registry Study Finds

Among the small fraction of patients who undergo redo surgery for a failed TAVR device, outcomes are worse than for those receiving initial surgical aortic valve replacement, according to registry data.

Perhaps due to the increase in transcatheter valve-in-valve cases, repeat surgery for a failed bioprosthetic valve has been declining in recent years and newer data has shown better outcomes with the former. Nevertheless, surgery remains an option for some patients, though data are limited.

“Our results point to the fact that SAVR after early failure of TAVR is a complex, technically-demanding procedure and that's associated with really long operative times, increased perioperative morbidity and much higher than expected operative mortality than compared with redo SAVR operations,” lead author Oliver Jawitz, MD (Duke University Medical Center, Durham, NC), told TCTMD. “We really need to focus further research efforts on refining operative techniques in performing TAVR explants and sharing best practices when these devices fail. That's going to become even more important in the future as younger and lower-risk patients undergo TAVR who might be candidates for surgical valve replacement after their valves fail.”

The paper was published online today ahead of print in JACC: Cardiovascular Interventions.

Higher Operative Mortality in All

For the study, Jawitz and colleagues included the 123 patients (median age 77 years) from the Society of Thoracic Surgeons (STS) adult cardiac surgery database who underwent surgical aortic valve replacement after receiving TAVR between 2011 and 2015. In total, 17%, 24%, and 59% had STS predicted rates of mortality of less than 4% (low), 4-8% (intermediate), and greater than 8% (high), respectively. Most patients (81%) had heart failure symptoms within 2 weeks before their operation.

SAVR was performed a median of 2.5 months following TAVR, and the most common reasons for needing a second procedure were paravalvular leak (15%), structural prosthetic deterioration (SVD; 11%), failed repair (11%), sizing/position issues (11%), and prosthetic valve endocarditis (10%). Simultaneous mitral valve replacement and CABG were performed in 2% and 6% of patients, respectively.

The operative mortality rate—death within 30 days of procedure or before hospital discharge—was 17.1%, and the observed versus expected mortality ratios were heightened regardless of baseline mortality risk (low 5.48; intermediate 1.66; high 1.16). The authors point out that overall operative mortality for a population of patients undergoing isolated SAVR between 2011-2013 was only 4.6%. Operative mortality was higher for patients with endocarditis, paravalvular leak, sizing/positioning issues, or failed repair compared with those who required reoperation for prosthetic deterioration.

We really need to focus further research efforts on refining operative techniques in performing TAVR explants and sharing best practices when these devices fail. Oliver Jawitz

Interestingly, 14% of patients required a second reoperation following SAVR within the same hospitalization, and 41% needed more than a day of ventilation. Discharge to home was only done in 43% of patients, while most of the remainder were discharged to an extended care facility or transitional care unit.

“We anticipated that surgical valve replacements in patients with a prior TAVR would be a higher-risk operation than maybe somebody undergoing a first-time surgical valve replacement, but probably not as high of a risk as somebody who has had a surgical valve replacement before and was undergoing a redo procedure,” Jawitz said. “We anticipated that the risk would be somewhere between the two. And we actually ended up finding that the risk associated with SAVR after TAVR was even higher than patients who had undergone a surgical valve replacement in the past. So, we were quite surprised by that.”

Plans for Longer-term Follow-up

In an accompanying editorial, Thomas MacGillivray, MD, and Michael Reardon, MD (both Houston Methodist Hospital, TX), point out that this study only included 0.3% of the roughly 40,000 TAVRs done during the study period, an underestimation, they say, “given the STS data only captures patients who had surgery.” Additionally, the data indicate that the included cases are “technically more challenging than a standard SAVR, taking longer to do and having a higher complication rate and early mortality.”

Because of this, “valve ingrowth and any difficulty in removing the TAVR valve cannot be ascertained . . . but remains a key question,” they write. “These are early and very unusual procedures to which these outcomes apply. It remains unknown if the same results will be found for SAVR after TAVR in the longer term when biologic SVD is generally seen. The mortality for SAVR after TAVR in this report is three times higher than the operative mortality for reoperative SAVR in multiple large series. Before embarking on a strategy of TAVR in younger low-risk patients, consideration of the risks of morbidity and mortality of subsequent procedures should be part of the shared decision-making.”

Predicting valve behavior in the long term has always been difficult and a retrospective, observational analysis can raise as many questions as it does answers, MacGillivray and Reardon say. Although this study adds to the knowledge base, they continue, “before we move to endorsing a TAVR to SAVR to TAVR as a potential pathway for young patients we need to know how difficult it will be to separate the TAVR from the aortic complex and how this will affect the conduct and risk for SAVR.”

Jawitz said he agreed the small size of the study and its use of older data limit the results. That’s why it is “going to be important to over time keep taking a look at this data to see [how these results change] as we as a community become more comfortable with operating on these patients,” he countered.

It’s likely that “a big wave of patients who require subsequent interventions” will come in the near future as TAVR moves to younger and lower-risk patients, he continued. “The morbidity and mortality associated with those procedures might be different than what we're seeing here, and so it's definitely going to be important to relook at this data down the road when we have more patients who've had their TAVR valves for longer.”

  • Jawitz OK. Gulack BC, Grau-Sepulveda MV, et al. Reoperation after transcatheter aortic valve replacement: an analysis of the Society of Thoracic Surgeons database. J Am Coll Cardiol Interv. 2020;Epub ahead of print.

  • MacGillivray TE, Reardon MJ. Reoperation after transcatheter aortic valve replacement: breaking up is hard to do. J Am Coll Cardiol Interv. 2020;Epub ahead of print.

  • The Society of Thoracic Surgeons (STS) National Database provided the data for this research.
  • Jawitz received funding provided by the NIH.
  • MacGillivray and Reardon report no relevant conflicts of interest.