Surgical Bailout Rarely Needed Though Can Save Lives: TVT Registry

The study adds new fuel to the debate over whether surgical backup is necessary at TAVR sites.

Surgical Bailout Rarely Needed Though Can Save Lives: TVT Registry

Surgical bailout during TAVR arises in about one out of 100 cases, with half of all patients whose procedures are converted to surgery surviving as a result, according to US registry data. The study adds more information to the debate over whether surgical backup should be a requirement for all TAVR cases.

Right now, TAVR in the United States is only reimbursed at sites that have both an interventionalist and a surgeon as part of the heart team, but as the procedure has been simplified over the years, many cardiologists have questioned the surgeon’s role, particularly in elective, low-risk cases.

The results definitively show that surgical bailout saves lives, lead author Andres Pineda, MD (University of Florida College of Medicine Jacksonville), told TCTMD.

“You can see it either way around: the glass half empty or half full. Fifty percent of those patients are saved, so that’s a pretty good argument,” he said, noting that prior studies looking at TAVR patients who have been converted to open surgery have been small and often have not followed patients to look at out-of-hospital outcomes.

“We were exactly at this point with PCI a few decades ago. It’s been a similar progression,” Pineda explained. “At this point we probably have enough to do a randomized study to look at this to settle the question [in TAVR]. I think that’ll be the only way. I think the surgical colleagues should still be involved in decision-making—they are very important—but this could lead to, maybe, a change.”

Tsuyoshi Kaneko, MD (Brigham and Women’s Hospital, Boston, MA), who was not involved with the study, told TCTMD these findings are not surprising and stem from advances in the field over time.

“The results are encouraging that surgical bailouts have been decreasing over time and I think it’s promising that we are accomplishing that,” he said. “It’s the combination of us understanding more from the days of using transesophageal echo as the sizing guide to now using CT scans as the gold standard. I think we are more careful about coronary obstructions and having [left ventricular outflow tract] calcium with the risk of rupture. I think we’ve learned it the hard way, but we’ve all gained the knowledge of these complications as an entire group and learned based on that.”

That said, “I don’t think this paper will add any further [support to] allowing a surgical nonpresence during TAVRs,” Kaneko argued. On the contrary, “it makes us think that surgeons being involved in these TAVRs on the heart team has a huge benefit because if something were to happen, there are lives that can be saved with cracking open the chest . . . in an emergency situation. So I don’t think the presence of surgeons can be underestimated.”

Moreover, Kaneko said a randomized trial in this space would be ”very tough” to do. “If you have a family member that is participating in a trial, would you ever want to be randomized in a hospital that does not have cardiac surgery backup? I don’t think randomization will be attractive for the patients,” he commented.

Surgery Saves Half

For the study, published in the September 23, 2019, issue of JACC: Cardiovascular Interventions, Pineda and colleagues looked at 47,546 patients who underwent TAVR at 396 US sites as part of the Society of Thoracic Surgeons/American College of Cardiology TVT Registry between November 2011 and September 2015.

A total of 558 patients (1.17%) needed conversion to open heart surgery for reasons including ventricular rupture (19.9%), prosthetic valve dislodgement into the left ventricle (19.4%), annular rupture (14.2%), aortic dissection (8.2%), and coronary occlusion (6.1%). Bailout patients were more likely to be older, female, smaller, and treated with nonfemoral access.

Almost half of patients in the bailout surgery group died in the hospital (49.6%) compared to only 3.5% of patients who had a successful TAVR (P < 0.0001). Also, except for new permanent pacemaker implantation, rates of other in-hospital nonfatal complications like MI, stroke, and bleeding as well as MACE (all-cause mortality, MI, and stroke) were higher for those who required surgical conversion.

Among the 68.9% of the cohort with 30-day and 1-year data available, both all-cause mortality and MACE rates remained significantly higher for those who underwent bailout surgery compared with those who did not. Nonfatal complications also were equivalent.

Outcomes Among Patients With Follow-up


Bailout Surgery

(n = 388)

Successful TAVR

(n = 32,370)

P Value

All-Cause Mortality





30 Days



< 0.0001


1 Year



< 0.0001






30 Days



< 0.0001


1 Year



< 0.0001


Notably, the incidence of surgical bailout decreased significantly over the study period (P = 0.0088), as did the use of nonfemoral access (P < 0.0001).

On multivariate analysis, the independent predictors of need for surgical bailout during TAVR were female sex, increasing hemoglobin, increasing LVEF, cardiogenic shock or left ventricular assist device use, salvage procedures, and nonfemoral access. The only predictor of survival following surgical bailout was increasing body surface area.

Pineda told TCTMD that some of these predictors like LV function and hemoglobin are “surprising and hard to put together with the clinical picture. . . . There are some theories there, but it’s hard to put together clinically.”

However, the absence of predictors coupled with an elective, transfemoral case might enable operators to “feel safe” performing certain TAVRs in the cath lab as opposed to an operating room, he said.

Surgical Backup Necessary?

In an accompanying editorial, Fabian Nietlispach, MD, PhD, Osmund Bertel, MD (HerzGefaessZentrum, Zurich, Switzerland), write that the study “has the potential to change practice in several ways.”

“With a 50% in-hospital mortality rate reported in these patients, the potential for improving mortality with on-site surgery is about 0.5%,” they explain. “From a statistical perspective and given the fact that patients can now be transferred with mobile biventricular assist devices to nearby hospitals (as suggested for complications arising during left atrial appendage closure), surgical standby probably is no longer a justifiable requirement to perform TAVR. From an individual standpoint, however, many of us would opt for surgical stand-by if we ourselves needed TAVR.”

I don’t think the presence of surgeons can be underestimated. Tsuyoshi Kaneko

This will “be a tough wall to crack unless there’s mounting evidence proving that [TAVR without surgical backup] is safe and valid to undergo,” Kaneko added. “The stakes will be high, even higher for the low-risk patients because these are the patients that are supposed to survive.”

Debate aside, the results should serve as a reminder to physicians to discuss surgical bailout with patients and families prior to a TAVR procedure. “Thoughtful patient selection is an issue not only for TAVR itself but also for surgical bailout when complications are met,” Nietlispach and Bertel write.

With even more refined patient selection, in addition to improvements in TAVR valves and procedures over the course of the study period, it’s possible that the bailout rate will drop below 1.17% in the years to come, they add. “Now that TAVR is indicated for the clear majority of patients with severe aortic stenosis, outcome research from real-world registries such as this study by Pineda et al is extremely valuable to realize the potential to further improve the results of TAVR as a breakthrough interventional treatment.”

Note: Several co-authors are faculty members of the Cardiovascular Research Foundation, the publisher of TCTMD.

  • Pineda reports receiving consulting fees from Pfizer and TZ Medical.
  • Nietlispach reports serving as a consultant to Abbott, Edwards Lifesciences, and Medtronic.
  • Bertel reports no relevant conflicts of interest.
  • Kaneko reports serving as a subcommittee member for the TVT registry.