Two Small Series Support Role for Cerebral Protection in TAVR

LONDON, England—Real-world use of cerebral embolic protection in patients undergoing TAVR for severe symptomatic aortic stenosis lowers the rate of periprocedural stroke when compared against TAVR procedures performed without it, according to the early results from a single academic medical center.

Benjamin Stripe, MD (UC Davis Medical Center, Sacramento, CA), who presented results from their first 100 patients treated with the Sentinel cerebral embolic protection system (Boston Scientific), reported they observed no periprocedural strokes when using the device. In contrast, there were five strokes and one TIA in the last 100 patients who underwent TAVR prior to implementing cerebral protection at their center (P = 0.02 for comparison).

“We looked at this because despite all the improvements in the technology and our techniques, TAVR is still associated with stroke,” Stripe said during a presentation at PCR London Valves 2019. He conceded, however, that “cerebral embolic protection still has a spotty record when it comes to the randomized trials.”

In the CLEAN-TAVI trial, for example, there was a reduction in the number of new brain lesions on MRI at 2 and 7 days, as well as a reduction in new lesion volume, but no changes in clinical outcomes with cerebral protection. A propensity-matched analysis showed a reduction in periprocedural stroke with cerebral protection compared with unprotected patients, but at the expense of increased major bleeding, said Stripe. Finally, the pivotal SENTINEL trial, which was presented at TCT 2016 and led to the approval of the device for capturing and removing thrombus/debris during TAVR procedures, did not show a clinical benefit, with investigators reporting no difference in the study’s primary endpoint of new brain lesion volume on MRI, or in stroke, compared with usual care.

While UC Davis Medical Center adopted the routine use of cerebral protection in all patients after the FDA approval in 2017, Stripe said default cerebral embolic protection during TAVR remains a hot topic given the absence of solid data supporting its use. Just because physicians believe in cerebral protection doesn’t mean it shouldn’t be studied, he said.   

We’ve been approached about being involved in a randomized trial, but we really hemmed and hawed about whether we wanted to randomize patients. Benjamin Stripe

“It rationally makes sense,” Stripe told TCTMD. “We catch stuff that’s visible to the naked eye in the vast majority of the filters. That’s not to say that we don’t think randomized data is important. We’ve been approached about being involved in a randomized trial, but we really hemmed and hawed about whether we wanted to randomize patients [in whom] we’d normally place a device. It gave us a lot of pause. Right now, though, we concluded we need the data. Even though it makes us feel rather uncomfortable, we’re going to enroll. We were doing TAVR for 7 or 8 years without [Sentinel], but now all of a sudden we have it and we’re very nervous to not use it.”

In the 100 patients treated with cerebral protection—the majority were at intermediate risk for surgery—there was one stroke between patient discharge and 45 days but no events after discharge among patients not treated with embolic protection (P = 0.06 for comparison at 45 days). There were no vascular complications with the device, but fluoroscopy time increased from 15.9 minutes to 19.7 minutes with cerebral protection (P = 0.003). Fluoroscopy dose also increased with use of the device.

Italian Experience With Dual-Filter Embolic Protection

In a second study, Tommaso Piva, MD (Ospedali Riuniti Torrette Di Ancona, Italy), also presented his institution’s experiences with cerebral protection, describing 195 consecutive patients at intermediate-to-high risk for surgery treated over 20 months.

Like the US experience, there were no periprocedural strokes or TIAs during TAVR procedures with the Sentinel device deployed. Visible debris was seen in 81% of treated patients, with physicians capturing emboli > 1 mm in 39% of cases. Speaking during his presentation, Piva said deploying cerebral protection takes very little time, with physicians able to get the device into position in less than 2 minutes in roughly 75% of cases. On average, the implantation time was 4 minutes.

“Is it well-wasted time or is it a waste of time?” he asked. Answering his own question, Piva said the few minutes required for deployment is a small price to pay for cerebral protection, noting that their center now uses embolic protection during all TAVR cases. He added that stroke is “just the tip of the iceberg,” and their series did not include any MRI measurements, neurocognitive evaluation, or histologic assessments.

Stripe said there is a rather short learning curve with the device and suspects the longer fluoroscopy time and higher fluoroscopy doses in their series can be lowered. “The quick learning curve is really quite remarkable,” he said. “If we tried to put the device in, we got the device in. It works in a remarkable number of different anatomies.” In their initial experience, just five patients were screened out of the first series of 100 patients because of inappropriate anatomy, such as severe tortuosity in the right subclavian artery.

Is There a Way to Stratify Patients?

To TCTMD, Stripe said there is no adequate means of identifying patients who might be at risk of stroke during TAVR, but that the move into lower-risk TAVR patients puts a greater onus on cerebral protection, even though these patients might be at a lower risk of stroke compared with older, higher-risk patients.

“We’re taking a younger, healthier population into a procedure that has a risk of stroke,” he said. “The potential for the loss of good quality of life is even higher in younger patients than in the older higher-risk population.” The downstream complications for a relatively young, active 68-year-old to have a disabling stroke would be more catastrophic than they would be in a 95-year-old patient who might not be as energetic, he said.

Cost is certainly an issue, said Stripe, but at their center they have not had difficulty with reimbursement from payers. In their initial 100 patients, he believes the additional cost of using cerebral protection, roughly $3,000, was offset by the reduction in clinical events. “If you had a bean counter sit down and just look at our first 100 patients, we had zero periprocedural strokes,” he said. “So zero patients in the ICU for 3 days and no extra neurological examinations. I think that in those first 100 patients, we probably made our money back even though we weren’t getting reimbursed for it.”

In 2018, the Centers for Medicare & Medicaid Services approved a New Technology Add-on Payment (NTAP) for Sentinel that included a maximum added payment of $1,400 for the fiscal year 2019.  

At his center, Piva said hospital administrators signed off on cerebral protection because they were convinced that capturing debris/emboli during the procedure was beneficial for patients. When patients are informed there is no strong evidence to support its use, at least with respect to preventing stroke, many still prefer to be protected, said Piva.

During the US Food and Drug Administration advisory panel hearing that led to approval, some committee members also went with their intuition, noting that it made sense to remove thromboembolic debris from circulation with Sentinel. At PCR London Valves, some experts made a similar case. “I’m a fan of cerebral protection,” said session moderator Nicola Buzzatti, MD (Ospedale San Raffaele, Milan, Italy). “I would want it used if I was undergoing TAVR.”