Coronary Obstruction From TAVR in Native Aortic Stenosis: Development and Validation of Multivariate Prediction Model

Area of Science:

• Cardiology
• Interventional Cardiology
• Medical Imaging

Background:

• Transcatheter aortic valve replacement (TAVR) is a key treatment for aortic stenosis.
• Coronary artery obstruction is a rare but serious complication of TAVR, associated with high mortality.
• Current methods for predicting this complication are insufficient.

Purpose of the Study:

• To develop and validate a predictive model for TAVR-related coronary artery obstruction.
• To identify anatomical predictors of coronary obstruction using preprocedural imaging.

Main Methods:

• Retrospective analysis of preprocedural computed tomography (CT) and fluoroscopy images.
• Development of a multivariate prediction model based on CT-derived measurements.
• Validation of the model against a propensity-matched cohort.

Main Results:

• Sixty patients experienced TAVR-related coronary obstruction; 1,381 did not.
• The obstruction cohort had significantly smaller annular and aortic root dimensions.
• The novel multivariate model demonstrated high predictive performance (AUC 0.93-0.94) for coronary obstruction.

Conclusions:

• A novel, CT-based multivariate prediction model effectively identifies patients at risk for TAVR-related coronary obstruction.
• This model can be integrated into routine clinical practice to enhance TAVR safety.
• Accurate prediction allows for timely preventive strategies and improved patient outcomes.