Transcatheter aortic valve replacement (TAVR) is a non-inferior alternative to surgery in the treatment of severe symptomatic aortic stenosis in high-and intermediate-risk patients and is now also approved for the treatment of severe symptomatic aortic stenosis in low-risk patients. Although the incidence of stroke following TAVR has decreased over the years ( Table 1 ), it remains a feared complication of the procedure. Stroke increases short-term and mid-term mortality following TAVR with significant morbidity in those that survive. In this issue of Cardiovascular Revascularization Medicine , Takagi et al. reviewed the predictors of peri-procedural and sub-acute cerebrovascular events (CVE) following TAVR in the Optimized transCathEter vAlvular interventioN-Transcatheter Aortic Valve Implantation (OCEAN-TAVI) registry. This is a welcome addition to the body of literature on this topic albeit with some flaws. They retrospectively reviewed 1613 consecutive high-risk patients with severe symptomatic aortic stenosis who underwent TAVR from October 2013 to July 2016 with the SAPIEN XT ( n = 1328) or SAPIEN 3 ( n = 141) balloon-expanding prosthesis (Edwards Lifesciences, Irvine, CA, USA) or the CoreValve ( n = 144) self-expanding prosthesis (Medtronic, Minneapolis, MN, USA). The TAVR procedure was performed using transfemoral, transiliac, transapical, trans-subclavian, and transaortic approaches, and prosthesis size was determined with pre-procedural imaging with echocardiogram and computed tomography scan. Heparin was administered to achieve an activated clotting time of greater than 250 s. All patients were observed in the intensive care unit post-procedure and they were continued on dual antiplatelet therapy for 3 to 6 months followed by aspirin indefinitely. For patients with clinical indication for anticoagulant therapy or who had high bleeding risk, the antithrombotic regimen was at the discretion of the operator. Suspected neurologic events were adjudicated by an on-site neurologist. Cerebrovascular events (CVEs) included transient ischemic attack (TIA) and clinical stroke (ischemia or hemorrhagic). Disabling stroke was defined using the modified Rankin Scale according to VARC-2 criteria. The study objective was to identify the predictors of peri-procedural and sub-acute CVEs following TAVR. The authors found that the 24-hour and 30-day CVE rates in the study cohort were 1.2% and 2.7%, respectively. They performed a multivariate analysis, which showed that indexed aortic valve area (iAVA) was the only independent predictor of 24-hour CVEs with a receiver operator curve-derived cut-off value of 0.40 cm ² /m ² for the prediction of 24-hour CVEs (adjusted OR 0.001; 95% CI, 0.001–0.13; p = 0.005). They also found that the independent predictors of 1- to 30-day CVEs were paroxysmal atrial fibrillation (PAF) (adjusted-OR, 3.35; 95% CI, 1.36–8.27; p = 0.009) and post-TAVR iAVA (adjusted-OR, 0.11; 95% CI, 0.02–0.66; p = 0.02). Prior ischemic stroke (adjusted-OR, 2.18; 95% CI, 1.07–4.45; p = 0.03), PAF (adjusted-OR, 2.18; 95% CI, 1.05–4.56; p = 0.04), and prior coronary artery disease (adjusted-OR, 1.88; 95% CI, 1.01–3.48; p = 0.05) were found to be independent predictors of 30-day CVEs.