NEW ORLEANS — The first embolic protection device designed to prevent periprocedural debris from entering any of the three cerebral vessels or the descending aorta during transcatheter aortic valve replacement (TAVR) was found noninferior to a device approved for the same indication.

Results from this randomized study showed the novel device, the Emboliner Embolic Protection System (Emboline), was about three times more effective for capturing procedural debris. These data appear to establish this device as “a next-generation” embolic protection system, according to Adam B. Greenbaum, MD, a co-director of the Structural Heart and Valve Center at Emory University School of Medicine in Atlanta.

“Current protection devices only partially protect the cerebral circulation,” said Greenbaum, providing the rationale for pursuing this technical advance over the system currently used.

The results of the study, called PROTECT H2H, were presented as a late-breaking trial at the American College of Cardiology (ACC) Scientific Session 2026.

Noninferiority Hard to Interpret

Although the study met its objective of noninferiority to the Sentinel Cerebral Protection System (Boston Scientific), the only device approved by the FDA for debris capture during TAVR, the result was difficult to interpret, said Michael N. Young, MD, director of the Structural Heart Disease Program at Dartmouth University’s Geisel School of Medicine in Hanover, New Hampshire.

Young, who served as a discussant for the trial, pointed out that the Sentinel device, to which the Emboliner system was compared, failed to show any significant benefit in two prior multicenter placebo-controlled trials, including BHF PROTECT-TAVI and PROTECTED TAVR, for hard endpoints, particularly stroke, when compared with no protection. He also noted that these cerebral protection devices are not guideline-recommended.

In this randomized but unblinded trial, the primary efficacy endpoint tested for noninferiority was a composite of death, stroke, or acute kidney injury. Stroke alone was one of two secondary endpoints. Debris capture, measured by number of particles, was the other.

The 522 primary TAVR patients randomized in the study had to be eligible for transfemoral access, which is required for the novel device. Exclusion criteria included a recent stroke and TAVR with a planned cardiovascular procedure, such as atrial fibrillation ablation or revascularization.

Eighteen centers in the US, Germany, and Brazil participated. Operators in all centers were experienced with the Sentinel device. Investigators were trained but not experienced with the experimental device, although Greenbaum maintained that neither is technically demanding.

At 30 days, 4.5% of those in the experimental arm vs 5.0% of those who in the Sentinel device arm had at least one event defining the primary composite endpoint, producing a P value for noninferiority of < .0004. There was no significant difference when results were compared for superiority.

Evaluated individually, all-cause death (2.0% vs 2.1%), stroke (2.0% vs 2.1%), and stage III acute kidney injury (0.4% vs 1.2%) along with major (2.0% vs 2.5%) and life-threatening (2.5% vs 3.3%) bleeding complications were numerically less frequent among those in the experimental arm. None of these differences reached statistical significance. There was also no significant difference in valve-related complications.

There were some significant procedural differences, including longer procedural time (72 vs 59 minutes; P = .001) and a longer fluoroscopy time (22 vs 19 minutes; P = .001), with the newer device. But Greenbaum argued that lack of prior experience relative to substantial experience with the Sentinel device is likely to have played a role in these differences.

Greater Debris Capture With Emboliner

For the secondary endpoint of debris capture, defined as the number of particles collected ≥ 150 µm, the mean count was 93 vs 31 (P < .0001) favoring the experimental device and produced a high degree of significance for superiority (P < .0001), Greenbaum reported.

In a detailed analysis of stroke events, Greenbaum noted that 7 of the 9 strokes occurred within 72 hours of TAVR, which is the period debris capture might be expected to provide protection against stroke. Of the two that occurred after 72 hours, there was one in each arm. In the periprocedural period, there were four strokes in the control arm and three in the experimental arm.

The fact that the study associated the newer device with significantly greater debris capture without showing significantly greater protection against events, particularly stroke, which Young described as “the holy grail” for proving the value of embolic protection devices, underscores the essence of the debate about these devices.

Young pointed out that stroke rates in recent trials have been under 1% and have been less than 2% in registries that might better reflect real-world data.

Data drawn from registries regarding the ability of embolic protection devices to reduce stroke rate have been mixed. Citing the largest set of data, which includes outcomes in more than 53,000 patients receiving embolic protection out of a total TAVR population of 400,000, Young noted that a reduction in disabling stroke was observed, but it was of borderline significance. This protection has not been seen in the two largest randomized trials, PROTECTED TAVR or BHF PROTECT-TAVI.

As a result, the clinical benefit of embolic protection devices has been hotly debated, Young said. While Greenbaum argued that even a small stroke risk reduction of 2% would provide a reasonable number-to-treat in favor of employing a protection device, Young remained uncertain if this level of stroke reduction can be achieved in the current era of TAVR safety.

Greenbaum reported having financial relationships with Abbott Vascular, Ancora Heart, Edwards Lifesciences, 4C, Gore Medical, JennaValve, Medtronic, Polares Medical, Transmural Systems. Young reported having financial relationships with Edwards Lifesciences and Medtronic.