Does valve in valve TAVR carry a higher risk for thromboembolic events compared to native valve TAVR?

Abstract Objective To evaluate the risk for brain lesions in valve in valve (VIV)‐transcatheter aortic valve replacement (TAVR) compared to native valve procedures (NV‐TAVR). Background TAVR is associated with the risk of stroke and a high incidence of brain lesions in diffusion‐weighted magnetic resonance imaging (DW‐MRI). The risk in VIV procedures is unknown. Methods Patients were included, if they received a Sapien‐3 or an Evolut‐R valve for an NV‐TAVR or VIV‐TAVR and a DW‐MRI within 3–5 days after the procedure. Results Two hundred nine patients had NV‐TAVR and 41 patients had VIV‐TAVR. The VIV group was significantly younger (mean 82.3 vs. 74.7 years old, p < .001) but had higher mean EuroSCORE II (4.8 vs. 7.8, p < .001). Predilatation was significantly more common with native valve (39% vs. 2%, p < .01), postdilatation rate was lower in VIV but without statistical significance (17% vs. 7%, p = .110). Combined in‐hospital safety outcome was comparable [13(6.6%) vs. 5(12.2%), p = .205]. Any stroke was documented in four (0.5%) patients after native valve and in one (2.4%) after VIV‐TAVR. New DW‐MRI brain lesions were significantly more often found in patients with native valves [153(73.2%) vs. 21(51.2%), p = .005], mean number of new lesions was also higher with native valve (2.8 ± 3.2 vs. 1.0 ± 1.4, p < .001). Forward stepwise of logistic regression found age and post dilatation significant predictors of new MRI lesions. Conclusions Compared to native valves, VIV procedures were associated with lower incidence and number of brain lesions according to DW‐MRI; these were explained by lower patients' age and lower rate of postdilatation.