Relationship between procedural characteristics and cerebrovascular events after transcatheter aortic valve replacement

Objectives The objective was to assess the impact of procedural characteristics on risk of stroke or transient ischaemic attack (TIA) after transcatheter aortic valve replacement (TAVR). Methods We included 370 consecutive patients who underwent balloon-expandable TAVR from 1 November 2008 to 30 June 2014. Procedural characteristics that may be associated with stroke/TIA were assessed. The primary outcome was stroke/TIA at 30 days. A propensity score was constructed using a logistic regression model with 29 parameters. Cox proportional hazards models were used with a propensity score covariate. Results Mean age was 80.9±7.9 years and mean Society of Thoracic Surgeons score was 8.3±5.0. The total number of balloon dilations ranged from 2 to 7. Out of 370 patients, 13 patients (3.5%) suffered stroke/TIA in the first 30 days after TAVR. In univariate analysis, postdeployment balloon dilation (PD) (HR 3.8, 95% CI 1.24 to 11.61; p=0.02) and emergent cardiopulmonary bypass (CPB) (HR 9.66, 95% CI 2.66 to 35.15; p<0.001) were significantly associated with 30-day stroke/TIA. In the multivariable Cox-proportional hazards model, PD (HR 4.95, 95% CI 1.02 to 24.03; p=0.04) and emergent CPB (HR 7.15, 95% CI 1.39 to 36.89; p=0.02) were independently associated with increased risk of 30-day stroke/TIA after adjusting for propensity score, total number of balloon dilations and periprosthetic regurgitation. Conclusion Postdilation as compared with total number of dilations, and emergent CPB were independently associated with increased risk of clinical neurological events in the first 30 days after TAVR. Reduction in balloon postdilation with appropriate valve sizing may reduce the risk of stroke or TIA after TAVR.