TAVR Without Balloon Aortic Valvuloplasty: Increasingly Common, With Similar Outcomes

It is now no longer necessary to perform routine balloon aortic valvuloplasty (BAV) before every TAVR case, and in fact, its use has been progressively declining over time, according to new observational data.

In the early days of TAVR, operators were encouraged to perform predilatation to facilitate easier delivery and help with device sizing. With more experience, some have been opting not to take this extra step, claiming that it adds unnecessary procedure time and potential increased risks of stroke, need for pacemaker, and acute aortic regurgitation.

“The necessity of systematic BAV has been debated over the last years and is now avoided by most of the teams,” write lead study author Pierre Deharo, MD (CHU Timone, Marseille, France), and colleagues.

Senior study author Thomas Cuisset, MD, PhD (CHU Timone), told TCTMD by email that his institution recently switched from “routine BAV to very selective use of BAV before TAVR,” opting instead for so-called “direct TAVR” in the majority of patients. Conducting this study was a means of determining whether this decision was justified.

Their results, published online ahead of print in JACC: Cardiovascular Interventions, show that many of their colleagues at other hospitals have made similar practice changes. Among 5,784 patients who underwent TAVR between 2013 and 2015 at 48 French centers, BAV was used in 55.4% of cases, but this rate progressively fell from 78.0% of cases in 2013 to 49.0% in the last trimester of 2015.

There were no differences observed between valve types as to whether BAV was performed; however, transfemoral access was more often used in patients who received BAV (86.1% vs 82.9%; P < 0.01). The primary endpoint of device success also did not differ between the direct TAVR and BAV cohorts (97.3% vs 97.6%; P = 0.40).

Direct TAVR with no BAV was associated with simpler procedures: fluoroscopy time (17.2 vs 18.5 minutes), radiation dose (608.9 vs 667.0 Kerma), and contrast load (131.5 vs 141.6; P < 0.01 for all) were all lower in this study arm, even after adjustment for other factors. Also, direct TAVR was associated with lower postoperative EF and similar aortic surface and mean gradient, but the rate of aortic regurgitation > 2 was higher in patients with predilatation (8.3% vs 12.2%; adjusted P < 0.01). There were no differences in stroke, permanent pacemaker implantation, acute kidney injury, or annular rupture between the study cohorts. Tamponade was more frequent in the BAV group (2.3% vs 1.5%; P = 0.04), a difference that was observed after adjustment for patient and procedural variables.

‘Individualized’ Decision

The notion that eliminating BAV would simplify the TAVR procedure “was obviously expected,” Cuisset said. “Conversely, reduction of PVL (observed mainly with balloon-expandable valves) was more surprising, even if many potential explanations can be proposed. The reduction of cardiac tamponade was also an important finding and coherent with the reduced number of manipulations and catheters with direct TAVR.”

Because of this and other similar studies, he believes the decision to perform BAV before TAVR should be individualized going forward. In his current practice, Cuisset added, less than 10% of transfemoral TAVR cases will include BAV prior to valve implantation and mainly because of some combination of “extreme aortic stenosis with heavily calcified valve, bicuspid valve in some cases, [and] major tortuosity, making valve crossing more challenging.”

Similarly, Samir Kapadia, MD (Cleveland Clinic, OH), who was not involved with the study, told TCTMD that his institution made the call 2 years ago to stop performing BAV on a routine basis prior to TAVR. He said there are only two indications that lead his team to perform predilatation now: difficulty crossing the valve with the catheter (often due to severe calcification or difficult anatomy) or trouble deciding on a valve size. “If it's a borderline size, then we use a balloon to size it,” he explained. “We put a balloon, inject dye, and size it. This also sometimes allows us to see if the coronaries will be impacted with the balloon.”

In an accompanying editorial, Shikhar Agarwal, MD (Geisinger Medical Center, Danville, PA), says the study addresses an important question for operators performing TAVR. “When posed with the question of predilation, most operators use clinical, anatomic, as well as functional information from the echocardiogram and multidetector computed tomography to decide if they would perform BAV prior to valve implantation. Clearly, the ones that require predilation represent the valves with heavy leaflet/ outflow tract calcification, which portends a higher grade of PVL post-valve implantation. Postimplant PVL may very well be a function of underlying valvular anatomy rather than predilation per se. It would be insightful to understand the predictors that led to predilation in this cohort of patients.”

Further, research has not yet shown whether valve type plays a role in outcomes after TAVR with or without BAV, Agarwal says. For paravalvular leak specifically—seen at a higher rate only among patients implanted with balloon-expandable valves in a prior study—“one potential mechanism that might explain this difference [is] that one-time inflation during direct TAVR with balloon-expandable valves cracks the valve and subsequently seals these cracks with one inflation. BAV prior to TAVR might crack the valve in different locations, which might not be sealed with the second balloon inflation for valve deployment, resulting in more PVL. This difference might not be appreciated with the self-expanding valves, potentially due to greater degree of PVL seen with these valves.”

Kapadia said his institution mainly uses balloon-expandable valves, “and for that reason not doing a predilation works for us.” A lot of sites that predominantly use self-expanding devices “also do not use predilation, and if necessary, they use postdilation. In my mind, that is better, that if you do postdilation, then you'll expand the valve better and you will not have the cracks that you will seal and that may be the reason, but this is a hypothesis.”

Randomized Data Await

While no randomized data yet exist on BAV before TAVR, Agarwal notes that the existing literature is “markedly heterogeneous at this point. The omission of BAV prior to valve implantation seems to simplify the procedure with potential reduction in postimplant PVL.” He mentions a few ongoing randomized trials that will seek to answer the question of need for postdilation as well as the possibility of increased stroke risk, including DIRECT, DIRECTAVI, and SIMPLIFy TAVI.

These trials will be important, but TAVR “devices have become so easy to cross without predilation that there is no need in my opinion for a randomized trial,” Kapadia said. “[There] is enough evidence to change our practice where we don't routinely need to predilate all the valves before doing the TAVR.” Still, he would like to see more information about stroke and need for pacemaker following direct TAVR in order “to understand the differences between balloon-expandable and self-expanding valves.”