Cerebral embolic protection device utilization and outcomes in transcatheter aortic valve replacement: A nationally representative propensity matched analysis

Abstract

IntroductionIn patients undergoing transcatheter aortic valve replacement (TAVR), cerebral embolic protection devices (CEPD) are used to possibly diminish the risk of periprocedural stroke. Trends and outcomes of CEPD usage in TAVR are not well characterized. MethodsNational readmission databases (NRD) 2017–2019 was used to identify hospital admissions for TAVR using ICD-10 codes, with versus without Sentinel CEPD. Primary outcomes of the study were in-hospital and 30-day stroke. Secondary outcomes include in-hospital mortality, 30-day mortality, 30-day readmission rate, and other procedural complications. We matched both cohorts using propensity score matching (PSM) and performed logistic regression to compute the odds ratios (ORs) and corresponding 95 % confidence intervals (CI). ResultsOut of 190,837 TAVR admissions in the United States, 10,643 (5.6 %) patients had TAVR with Sentinel CEPD. After propensity score matching, our cohort included 10,503 patients with CEPD and 10,541 without CEPD. Trends in CEPD utilization are noted in Fig. 1. In the PSM cohort, Sentinel CEPD was not associated with decreased risk of in-hospital stroke (1.9 % vs. 1.8 %, OR: 0.98, 95 % CI: 0.76–1.26, p = 0.88), 30-day stroke (2.1 % vs. 2.1 %, OR: 1.01, 95 % CI: 0.78–1.30, p = 0.96), or 30-day mortality (1.3 % vs. 1.0 %, OR: 0.74, 95 % CI: 0.51–1.07, p = 0.11) when compared to TAVR without CEPD. Other in-hospital and short-term outcomes post-TAVR were not impacted by Sentinel CEPD usage, including acute kidney injury, vascular complications, paravalvular leak, cardiogenic shock, circulatory support, or permanent pacemaker (Table 1). ConclusionIn this nationally representative cohort, Sentinel CEPD utilization during transfemoral TAVR for stroke prevention was not associated with reduced odds of in-hospital stroke, 30-day stroke, or 30-day mortality. Future studies should focus on optimizing patient selection for CEPD and establishing predictive models to identify the subset of TAVR patients with higher risk for periprocedural stroke who might benefit from CEPD.