Stroke Benefit Missing With TAVR Embolic Protection

Early real-world use of the Sentinel cerebral embolic protection device did not reduce procedural strokes after transcatheter aortic valve replacement (TAVR) for severe symptomatic aortic stenosis, a study found.

The incidence of stroke or transient ischemic attack (TIA) was about the same whether patients received commercial TAVR with or without the Sentinel dual-filter system designed to capture embolic debris dislodged during the procedure (1.8% vs 2.2%, P=0.15), according to a group led by Mohamad Alkhouli, MD, of Mayo Clinic School of Medicine in Rochester, Minnesota.

Lack of improvement for Sentinel cases was also observed in the secondary endpoints of ischemic stroke, hemorrhagic stroke, fatal stroke, disabling stroke, death, length of stay, and cost, the authors reported online in a research letter in JAMA Internal Medicine.

The study "highlights the uncertainty regarding the effectiveness of EPDs [embolic protection devices] in stroke prevention during TAVR," noted Rohan Khera, MD, MSCS, of the University of Texas Southwestern Medical Center in Dallas, and Saket Girotra, MD, SM, of the University of Iowa and Iowa City Veterans Affairs Medical Center, writing in an invited commentary.

"The clinical significance of embolic debris and new brain lesions during intracardiac procedures is not firmly established and, to our knowledge, no subsequent RCT [randomized clinical trial] of the EPD to determine its efficacy on stroke prevention has been pursued," Khera and Girotra said.

Sentinel remains the only embolic protection device available in the U.S. following its 2017 de novo class II device approval for use during intracardiac procedures. Such classification allows for the device to be used in TAVR centers of excellence across the country.

Approval had been based on the SENTINEL trial in which retrieved devices filtered out histopathologic debris in 99% of patients but failed to significantly reduce new lesion volume in protected brain territories or reduce major adverse cardiac and cerebrovascular events.

"While the FDA labelling is consistent with the trial data, it is somewhat misleading because the EPD was not explicitly cleared for stroke prevention, the intended use of the device," Khera and Girotra said.

One possibility for why the Sentinel hasn't yet been associated with a robust clinical benefit, they speculated, may be the lack of protection of the left vertebral artery.

The present observational study was based on the Vizient database that captured 10,985 TAVR cases in 2017 and 2018. Participating sites were more than 400 U.S. academic centers and their affiliated hospitals.

Patients were two groups who got TAVR in non-overlapping time periods: people who underwent Sentinel protection from July 2017 to Dec. 2018 (since FDA approval of the device) and those who had unprotected TAVR from Jan. to June 2017.

The results were the same before and after propensity score matching to adjust for baseline differences (such as patients in the unprotected group presenting with a higher prevalence of key comorbidities but having similar rates of carotid stenosis and obesity as their protected peers).

The current study would therefore appear to contradict smaller observational studies suggesting a benefit in stroke prevention for the Sentinel.

Alkhouli's team reported that the proportion of hospitals using the Sentinel in TAVR increased from 8.6% in the third quarter of 2017 to 32.4% in the fourth quarter of 2018. The proportion of protected TAVR patients also grew from 2.8% to 17.3%.

Limitations of the study, the commentators said, include its reliance on ICD codes for defining stroke, a lack of detailed clinical and procedural information for matching, and the use of non-contemporaneous controls.

"It may be argued that patients receiving EPDs in the study by Alkhouli et al had a higher preprocedural stroke risk and the lack of benefit represents unmeasured confounding," Khera and Girotra wrote. "However, in the absence of criteria for selective use and the large hospital variation in device use, the nearly identical rate of stroke/TIA in patients who received or did not receive EPDs suggests that the lack of difference is unlikely to reflect residual confounding."

Ultimately, the story of early Sentinel use is reason for caution against the reliance of surrogate endpoints and observational studies, the commentators concluded. "It is imperative that regulatory clearance for clinical devices is based on unequivocal evidence of patient benefit. In the case of cerebral EPDs, efficacy in preventing strokes, not just capturing embolic debris, should be the evidence standard before unrestricted clinical use."