Rate of peri-procedural stroke observed with cerebral embolic protection during transcatheter aortic valve replacement: a patient-level propensity-matched analysis.

Aims: The role of cerebral embolic protection (CEP) in transcatheter aortic valve replacement (TAVR) remains controversial. Randomized trials have not been powered to demonstrate a reduction in stroke rates. The aim of this patient level pooled analysis was to validate the impact of the dual-filter CEP device (Claret Medical Inc., CA, USA) on peri-procedural stroke in a large number of TAVR patients.
Methods and Results: Patients from the SENTINEL US IDE trial were combined with the CLEAN-TAVI and SENTINEL-Ulm study in a patient level pooled analysis (N = 1306). Propensity score matching was performed to adjust for possible confounders. The primary endpoint was procedural stroke within 72 h post-TAVR according to Valve Academic Research Consortium-2 criteria. The secondary endpoint was the combination of all-cause mortality or all-stroke within 72 h after TAVR. In the propensity-matched population, 533 patients underwent TAVR without CEP and 533 patients underwent TAVR with CEP. TAVR patients without vs. with CEP were similar with respect to baseline characteristics, procedural approach, or valve type. In patients undergoing TAVR with dual-filter CEP, procedural all-stroke was significantly lower compared with unprotected procedures [1.88% vs. 5.44%, odds ratio 0.35, 95% confidence interval (CI) 0.17-0.72, relative risk reduction 65%, P = 0.0028]. In addition, all-cause mortality and all-stroke were significantly lower (2.06% vs. 6.00%, odds ratio 0.34, 95% CI 0.17-0.68, relative risk reduction 66%, P = 0.0013).
Conclusion: Our findings suggest that TAVR with the dual-filter CEP device is associated with a significant lower rate of peri-procedural stroke compared with unprotected procedures. However, randomized trials are still needed to clarify this issue.
(Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com.)