Cerebral Protection in TAVR: New Insights, New Questions as First Device Seeks FDA Clearance

LONDON, England—Results for the first US-based randomized controlled trial testing use of a cerebral protection device in TAVR are due out next month, but in Europe, where a number of devices already hold CE Mark approval, TAVR operators speaking this week at PCR London Valves say they are using cerebral protection in “most” cases.  

Whether US operators will follow suit depends on how the Food and Drug Administration (FDA) views the incoming data. On Tuesday, in what may be an early sign of results scheduled to be presented next month as a late-breaking trial at TCT 2016, Claret Medical announced it had filed with the FDA for clearance of its Sentinel cerebral protection system.

During London Valves, Michele Dallago, MD (University of Padua, Italy), presented results of a small series of cases using the Sentinel. The device is made up of two independent filters and is deployed via the aortic arch—the larger proximal filter is deployed in the brachiocephalic trunk, and the smaller distal filter is deployed in the left common carotid artery.

Of 22 patients in whom Dallago and colleagues were able to successfully deploy the device (in two patients, tortuosity of the brachiocephalic artery prevented placement), macroscopic debris was captured in 17 cases (77%). There were no strokes or TIAs within the first 48 hours postprocedure, no bleeding or vascular complications, and no stroke/TIA within 30 days post-TAVR.

The debris captured—Dallago showed the always-provocative photos of the heavily laden filters—as well as the safety and feasibility of the procedure has prompted him to use filter protection in “all” cases, he said following his presentation.

Those With the Option Take It

Where healthcare funding models permit, other European operators have adopted a similar stance, based largely on results from the European CLEAN-TAVI trial, first presented at the TCT 2014 meeting and subsequent analyses and registry data. As reported by TCTMD, data presented earlier this year at EuroPCR showed that embolic protection during stroke appeared to reduce the incidence of central nervous system infarction as well as the average diffusion-weighted MRI volume and total volume at 30 days.

Equally galvanizing have been studies showing the amount of debris captured in the filters and deflection devices. Again, a separate study of 217 patients, also presented at EuroPCR, showed that that thrombotic debris following TAVR was found in 99% of cases.

Back in May, those findings prompted session moderator Raj Makkar, MD (Cedars-Sinai Heart Institute, Los Angeles, CA), to tell TCTMD that if he had access to cerebral protection devices, he would “be tempted to use them for everybody.” 

At London Valves, however, a number of expert panel discussions circled back to the fact that no study has definitively shown that cerebral protection reduces clinical events.

“If you are using it in all patients, it increases the complexity of the procedure as well as the cost and time” Jonathan Byrne, MD (London Bridge Hospital, England), commented following Dallago’s presentation. “We need more information as to the clinical event rate in patients who weren’t treated with cerebral protection before deciding how and when to use these devices.”

What to Measure, and When?

Following Dallago’s presentation, Masieh Abawi, BSc (University Medical Centre Utrecht in the Netherlands), presented the results of a pilot study designed to assess the effect of TAVR without cerebral protection on midterm cognitive function using a battery of pre- and postprocedure cognitive tests. These included the mini mental state, clock drawing, immediate recall recovery, delayed recall memory, an eight-word recognition test and trail-making tests.

Abawi reported that despite increased cerebral ischemic lesions on imaging, cognitive testing found no impairments at 3 to 6 months. Indeed, the 30 patients who underwent pre- and post-TAVR cognitive testing in the pilot actually showed significant increases in immediate recall memory after their procedure.

Speaking with TCTMD, Abawi emphasized that the results support those from the DEFLECT III trial, which also showed improvements in cognition after transcatheter aortic valve interventions. “That could be the effect of better ejection fraction, higher blood pressure after TAVI, better perfusion of the brain despite these cerebral ischemic lesions seen on imaging,” he explained. The problem, he continued, is that “we still don’t know the effect of these lesions in the long-term. As we’ve previously showed, these lesions affect the early outcome of TAVI, for example delirium.”

As reported by TCTMD, Abawi and colleagues have previously reported rates of delirium in more than 13% of 270 patients following unprotected TAVR and noted a link between delirium and longer hospital stays as well as increased mortality—the latter following transfemoral TAVR. During PCR London Valves, Abawi’s group published new data on delirium in 103 patients who also underwent postprocedure MRI showing a higher number of new cerebral diffusion-weighted MRI lesions (found in 90% of patients undergoing unprotected TAVR) among those who developed perioperative dementia (which occurred in 15% of patients). The new findings were published online this week as a research letter in the Journal of the American College of Cardiology.

Senior author Pieter Stella, MD, PhD, (University Medical Centre Utrecht), told TCTMD that his center now uses cerebral protection in all cases with “suitable anatomy”—approximately 90% of patients.

Delirium is not, however, listed on clinicaltrials.gov as being among the eagerly anticipated clinical endpoints for SENTINEL. Instead trialists led by Susheel Kodali (Columbia University Medical Center, New York, NY), and Samir Kapadia, MD (Cleveland Clinic, OH), are looking at a primary efficacy endpoint of reduction in mean lesion volume on diffusion weighted-MRI and primary safety endpoint of major adverse cardiac and cerebrovascular events at 30 days.

Should the trial prove positive, the next big questions will be whether to use the device in all patients, or just the as-yet unidentified high-risk patients. It’s also unknown whether cerebral protection is more necessary for certain types of TAVR devices but not others. SENTINEL is being conducted in the United States and Germany and is permitting the use of “all commercially available” TAVR devices—which in the US means the CoreValve (Medtronic) and Sapien (Edwards Lifesciences) devices. Contacted by TCTMD, Kapadia said investigators have not yet decided whether to present results broken down by device type “but we may be able to do so.” 

On this point, Stella told TCTMD that his group has not seen any differences between the different transcatheter valves in the amount of debris captured. Their center uses the TriGuard device (Keystone Heart), and they’ve used the Sapien, CoreValve, Evolut R (Medtronic), Direct Flow (Direct Flow Medical), and Acurate Neo (Symetis) TAVR devices.

HardTruths

If SENTINEL goes in the opposite direction—and shows a neutral or negative impact on brain lesions or safety—the results will be tough to swallow for operators already using the devices, encouraged by the amount of debris they capture and convinced this must be doing some good.

Speaking with TCTMD after Dellago and Abawi’s presentations, session moderator Alexandra Lansky, MD (Yale University, New Haven, CT), drew parallels with the thrombus aspiration/thrombectomy fields, where despite two large negative randomized trials, operators who see the captured thrombi during percutaneous coronary interventions have been reluctant to accept that the devices are at best useless and possibly even harmful. 

Whether thrombus capture in TAVR will follow the same path remains an open question. People see these clots extracted and think “they’ve done something,” Lansky said, but the clinical impact remains unproven. The problem, she continued, is that the debris entering the brain is a “diffuse barrage” that produces “heterogenous effects”—and figuring out how and when to measure this is difficult.

Lansky, for one, dismisses the parallels with the dashed hopes of thrombus aspiration, saying that you can’t compare the impact of thrombi in the coronary and cerebrovascular beds. “The brain is different,” she said.