May 29, 2015

PARIS, FRANCE — A porous cerebral-protection shield placed in the aortic arch during transcatheter aortic-valve implantation (TAVI) and retracted afterward seemed to reduce the 30-day burden of ischemic brain lesions associated with the procedure and improve some aspects of cognitive function compared with TAVI lacking such protection[1]. The apparent benefits in the small randomized DEFLECT 3 trial had also been evident at testing before patients left the hospital.

Although the 85-patient study's primary end point focused on safety and suggested no excess clinical in-hospital risk to TAVI using the TriGuard "embolic-deflection" device (Keystone Heart), analysis extending out to 30 days demonstrated new neurologic deficits in only one-fifth as many TriGuard-protected patients as in those who didn't get the device. Brain lesions were defined using diffusion-weighted MRI (DW-MRI) and the National Institutes of Health Stroke Scale (NIHSS).

The safety end point was seen in 21.7% of "protected" patients and 30.8% of controls (P=0.34) in hospital and in 26.1% and 31.2%, respectively (P=0.62), at 30 days, with no important between-group disparities in the end-point components, according to Dr Andreas Baumbach (University Hospitals Bristol, UK), who presented the study last week here at EuroPCR 2015. That safety end point encompassed all-cause mortality, all strokes, life-threatening or disabling bleeding, acute kidney injury, and major vascular complications.

A Role in All TAVI Procedures?

The TriGuard device, which allows blood to pass, but not potentially embolic debris, had undergone some design changes since earlier studies that made it easier to use in DEFLECT 3, Baumbach told heartwire from Medscape. It is "undergoing continued iterations for easier and more reliable positioning," he said, and "there is no safety signal suggesting implantation comes with a risk." In DEFLECT 3, the embolic-protection device was successfully put in place for the duration of the valve procedure in 40 of the 45 patients randomized to receive it.

Nor, apparently, does the debris it deflects away from the innominate artery and left common carotid and subclavian arteries cause issues further along the descending aorta. "It goes downstream. [There is] no signal to suggest there is a clinically relevant distal or kidney embolization," Baumbach said. And although the device requires a larger-caliber guide catheter than conventional TAVI, 9F instead of 6F, he noted elsewhere at the EuroPCR sessions, there was no sign of increased bleeding at the femoral access site.

"This device and the principle of embolic protection have a great potential to become a major part of adjunctive treatment during TAVI," Baumbach said to heartwire . "If safety can be demonstrated and the positioning of the device is very easy, I guess it should have a role in all TAVI procedures."

First Evidence It Reduces Ischemic Lesions

DEFLECT 3 was published in the European Heart Journal, with first author Dr Alexandra J Lansky (Yale University, New Haven, CT)[2], simultaneously with Baumbach's live presentation at the sessions.

"For the first time," according to the formal report, "it has been demonstrated that filter protection of the cerebral vessel takeoffs in the aortic arch is capable of increasing freedom from cerebral ischemic lesions on DW-MRI. Compared with controls, use of the TriGuard device during TAVI numerically reduced single and maximum lesion volumes."

In addition, as Baumbach reported at the sessions, DW-MRI disclosed ischemic lesions at 30 days that had not been apparent at postprocedure predischarge imaging in a similar 11.5% of TriGuard patients and 9.1% of controls.

"These are new lesions that happened after discharge, so by definition they cannot be prevented by a deflection device," according to Baumbach. But, he said, they suggest that there is an "ongoing embolic burden" after TAVI of about 10% out to 30 days.

In other new insights from the trial, patients who had TriGuard protection went on to have more thorough recovery of cognitive function than controls when measured at 30 days. The prevalence of cognitively unimpaired patients was similar in both groups at baseline and right after TAVI, based on Montreal Cognitive Assessment (MoCA) scores. But, from 20% in both groups after TAVI, it climbed significantly to 45.5% in TriGuard patients and not at all in controls by 30 days.

"What that means is, almost half of the patients who had protection had a normalized neurocognitive assessments score as opposed to only 20% in controls," Baumbach said.

Not Just Lesion Volume, But Location

Following Baumbach's presentation of the trial, session comoderator Dr Stephan Windecker (Universitätsspital Bern, Switzerland) asked about the gulf between stroke rates based on clinical outcomes vs imaging. "If you look at clinical-event rates in a contemporary series, it's about 3% to 4% for stroke, yet if you perform imaging it's nearly every patient who has lesions. How do you reconcile this large discrepancy? Are you too insensitive with the clinical observations or oversensitive with the imaging?" he asked.

Absolute changes in brain-lesion volume aren't the whole story, and a given embolism may be associated with "a large spectrum of clinical sequelae," observed Baumbach. "Of course it matters what volume of lesions you have in the brain, but it also matters where the brain lesion is—the location of the infarct is important for the clinical picture."

But what is clear from DEFLECT 3 and other TAVI trials, he said, "is that we cannot ignore these lesions. They may not produce overt stroke that we as cardiologists are able to detect, but they certainly produce neurological deficits that a neurologist can see, or our trained staff who [give the NIHSS test] can detect."

Better in One Valve Implant Than the Other

The trial enrolled patients in 2014 and early 2015 at 13 centers in Europe and Israel to TAVI with or without TriGuard protection; all had been referred for TAVI due to "high or extreme" surgical risk but had to be without recent MI, neurologic events, or shock.

Relative Risk (95% CI) for Safety End Point, TriGuard vs No TriGuard at TAVI in DEFLECT 3

Safety end points TriGuard (%) Control (%) RR (95% CI)
In-hospital MACCE 21.7 30.8 0.71 (0.34-1.46)
30-d MACCE 26.1 31.2 0.83 (0.37-1.84)
MACCE=all-cause mortality, all disabling or nondisabling stroke, life-threatening or disabling bleeding, stage 2 or 3 acute kidney injury, major vascular complications

New neurologic lesions, based on a combination of ischemia by DW-MRI and worsening NIHSS scores, were seen at discharge in 3.1% of TriGuard-protected patients and 15.4% of controls (P=0.16). "Though the between-group difference was not statistically significant in this trial, larger prospective studies should confirm a reduction in clinical neurological events if the observed discrepancy is maintained," according to the published report.

TAVI was performed via the femoral route in all but three cases and used the balloon-expandable Sapien valve (Edwards Lifesciences) in 63.5% of patients, the self-expandable CoreValve device (Medtronic) in 31%, and other transcatheter valves in the remainder.

The Sapien seemed to favor placement of the TriGuard device in DEFLECT 3, Baumbach told heartwire . "In this series it worked better with the balloon-expandable and steerable device. This can be steered through the aortic arch and will reliably stay below the filter. A nonsteerable device can take the outer curve of the arch and slightly displace the filter."

Acknowledging the preliminary nature of DEFLECT 3, Baumbach said an upcoming randomized trial called REFLECT, planned for the US and Europe, will look at "patient-focused end points" and "hopefully will document the efficacy of the device and the principle of embolic protection to improve patient outcomes in TAVI." Baumbach is coprincipal investigator for REFLECT, which is designed to be the pivotal trial for FDA approval of the TriGuard device.

DEFLECT 3 was supported by Keystone Heart and the National Institute of Health Research Bristol Cardiovascular Biomedical Research Unit. Baumbach discloses serving on a scientific advisory board for and receiving research grants from Keystone Heart.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.