乌鸦传媒

NEW YORK CITY -- Researchers eked out evidence of a small clinical benefit with the Sentinel cerebral embolic protection (CEP) device in transcatheter aortic valve replacement (TAVR) -- and pointed to the patients more likely to benefit -- on the basis of disabling strokes.

The incidence rate of in-hospital disabling stroke, defined as stroke associated with either in-hospital death or discharge to a non-home location, was 0.79% with the CEP device versus 0.91% without the device (RR 0.87, 95% CI 0.73-1.00) in nationwide STS/ACC TVT registry data from 2018 to 2023, the largest dataset to date in this field.

This reduction in disabling stroke in the real world was supported by a secondary analysis using propensity score weighting (OR 0.79, 95% CI 0.70-0.90), reported Neel Butala, MD, of Rocky Mountain Regional VA Medical Center and the University of Colorado School of Medicine, both in Aurora.

"Our prespecified primary analysis using an instrumental variable approach demonstrated a small, borderline significant reduction in in-hospital disabling stroke that is likely to be causal in nature," Butala said here at the New York Valves meeting, the successor to the annual TVT conference previously hosted by the Cardiovascular Research Foundation. The study was simultaneously published in .

Panelist Samir Kapadia, MD, of the Cleveland Clinic, called the study an "important contribution" that signals that the CEP field is still "active," with newer, better devices on the horizon.

Despite FDA approval of the Sentinel device, CEP remains controversial following multiple trial setbacks. The technology never entered mainstream practice because of what many consider to be an unproven benefit of capturing debris dislodged during TAVR.

For example, the Sentinel CEP device failed to reduce periprocedural strokes in the randomized PROTECTED-TAVR trial, even though it prevented embolic material from reaching the brain. In that study, the researchers tried and failed to find subgroup responders who would likely have strokes prevented by the CEP device.

Based on the TVT registry, however, Butala and colleagues did identify TAVR patients with prior stroke as a subgroup that disproportionately benefited from fewer disabling strokes with CEP.

"We've been trying to find this, a cohort of patients that would benefit more from EPD [embolic protection devices] and I think you kind of nailed it," said session discussant Alexandra Lansky, MD, of Yale School of Medicine in New Haven, Connecticut. "The sample size is large enough that patients with prior stroke are probably the ones we want to target."

There was no reduction in nondisabling strokes in the cohort overall or in any subgroup. This may be explained by the Sentinel device allowing smaller particles to get to the head and causing smaller strokes, Butala told the audience.

"These findings are congruent with those from the PROTECTED-TAVR trial and suggest that the reduction in disabling stroke from EPD use is real, but the effect size may be relatively small in all-comer patients," he concluded.

The investigators had conducted a retrospective study of first-time TAVR recipients in the STS/ACC TVT registry. They excluded emergent procedures, alternative access, and people who received a non-Sentinel CEP device.

In the end, there were 414,649 TAVR recipients included in the study (mean age 78-79, 39-44% women), of whom nearly 13% received a Sentinel device.

The Sentinel group had more bicuspid valves (8.9% vs 5.5%) and more prior aortic valve procedures (11.6% vs 8.1%) compared with controls. Otherwise, baseline characteristics were relatively well-matched, according to Butala's group.

Use of CEP devices had plateaued at 15% until the 3,000-patient PROTECTED-TAVR results were released in 2022 -- dropping down to 12% by mid-2023, Butala said. He highlighted the variation in CEP use across the study period, with 40% of sites reporting no CEP use and 6% using the devices for over half their cases.

The investigators acknowledged that their analysis was not randomized, but maintained that causal inference was supported by their quasi-experimental approach of an instrumental variable analysis with site-level preference for CEP use as the instrument.

Lansky added that the potential underestimating of event rates and lack of long-term mortality data were limitations to the study.

In addition, Butala and colleagues had relied on discharge location as a proxy for stroke severity, a variable that is not explicitly identified in the TVT registry. The disabling stroke endpoint was separately validated in a population pooling data from PARTNER 2A, PARTNER 3, the Sapien S3i Registry, SURTAVI, Evolut Low Risk, and PROTECTED TAVR -- with sensitivity of 80%, specificity of 66.7%, and a positive predictive value (PPV) of 58.2%.

"While we can be fairly certain that a patient discharged home after TAVR did not have a disabling stroke, the lower sensitivity and PPV temper our ability to verify if discharge to non-home location or death was truly due to a disabling stroke post-TAVR or due to another postprocedure complication(s) or for other reasons such as frailty," cautioned Dharam Kumbhani, MD, SM, of the University of Texas Southwestern Medical Center in Dallas, and Frederick Welt, MD, of the University of Utah Health Sciences Center in Salt Lake City, in an . "In addition, baseline residential status is not factored into this definition."

"The current endeavor by Butala and colleagues is a creative and interesting approach to try and shed light on a question that has yet to be definitively answered by randomized data," they added. "It potentially sets the stage for future studies and action items on this topic."

The duo pointed to the study as an ongoing randomized trial in the CEP space, but said it is unlikely to be powered for a disabling stroke endpoint.

For now, "we don't find evidence of harm. It's not like we're harming people using this device," Butala said, adding that the only downsides with the CEP device are longer procedure time and higher cost.