Evolution, Predictors, and Neurocognitive Effects of Silent Cerebral Embolism During Transcatheter Aortic Valve Replacement.

Abstract Objectives The aim of this study was to assess the characteristics, predictors, evolution, and neurocognitive effects of silent cerebral ischemic lesions (SCILs). Background Most patients undergoing transcatheter aortic valve replacement (TAVR) develop SCILs detectable on magnetic resonance imaging (MRI). The natural history and clinical relevance of SCILs are not well established. Methods Cerebral MRI was performed within 7 days before TAVR to assess baseline status and age-related white matter change score. MRI was repeated post-operatively to assess the occurrence, location, number, and dimensions of SCILs. Patients developing SCILs underwent a third MRI examination at 3- to 5-month follow-up. A neurocognitive evaluation was performed before TAVR, at discharge, and at 3-month follow-up. Results Of the 117 patients enrolled, 96 underwent post-procedural MRI; SCILs were observed in 76% of patients, distributed in all vascular territories, with a median number of 2 lesions, a median diameter of 4.5 mm, and a median total volume of 140 mm3. Independent predictors of SCIL occurrence were higher baseline age-related white matter change score and the use of self-expanding or mechanically expanded bioprostheses. Among 47 patients who underwent follow-up MRI, only 26.7% of post-procedural SCILs evolved into gliotic scar. SCIL occurrence was associated with a more pronounced transient neurocognitive decline early after TAVR and with lower recovery at follow-up. Conclusions SCILs occur in the vast majority of patients undergoing TAVR and are predicted by more diffuse white matter damage at baseline and by the use of non-balloon-expandable prostheses. Although most SCILs disappear within months, their occurrence has a limited but significant impact on neurocognitive function.