Background: There is conflicting data on stroke risk associated with self-expanding (SEV) compared to balloon-expandable (BEV) transcatheter heart valves (THV).

The aim of this study was to examine whether the risk of cerebral embolization assessed by diffusion-weighted magnetic resonance imaging (DW-MRI) is different between self-expanding versus balloon-expandable THVs.

Methods: Patients were included if they received an Evolut R SEV or a SAPIEN 3 BEV for the treatment of severe aortic stenosis without the use of cerebral protection systems and DW-MRI within 3-5 days after procedure. All patients underwent pre-procedural multi-slice computed tomography (MSCT) of the heart, the thoraco-abdominal aorta and the iliofemoral arteries. Two independent physicians assessed the aortic valve calcium complex volume (AVCCV). DW-MRI was analyzed for incidence, number, area, and volume of brain infarcts by a blinded investigator. Acute and thirty-day outcome was evaluated according to VARC-2-criteria.

Results: Out of 153 consecutive patients (82 ± 7 years, 55% female) 99 received a SEV, 54 a BEV. Baseline, echocardiographic and MSCT data did not differ between both groups. Pre- and postdilation was more often performed in the SEV-group (62 vs 22% and 41 vs 7.4%, respectively, p<0.001). The post-procedural number of brain lesions per patient was significantly higher in the SEV-group (3.4 ± 3.1 vs 2.2 ± 2.5, p=0.006). Logistic regression analysis identified SEV, age and AVCCV as predictors for the number of cerebral lesions.

Conclusion: In addition to age and AVCCV the THV-type is a significant predictor for the number of cerebral lesions after transcatheter aortic valve implantation.