Combining Coronary CTA Features with Clinical Risk Factors for Predicting Adverse Clinical Outcomes Following...

Area of Science:

• Cardiology
• Medical Imaging
• Predictive Analytics

Background:

• Transcatheter aortic valve replacement (TAVR) is a crucial procedure for aortic stenosis.
• Predicting adverse outcomes post-TAVR is essential for patient management.
• Coronary computed tomography angiography (CCTA) provides detailed cardiac imaging.

Purpose of the Study:

• To evaluate the predictive value of CCTA-derived features combined with clinical factors for adverse events after TAVR.
• To develop and validate a combined predictive model for post-TAVR outcomes.

Main Methods:

• A multicenter cohort of 406 TAVR candidates with pre-procedure CCTA was analyzed.
• Major adverse cardiac events (MACE) and all-cause mortality were primary and secondary endpoints.
• Cox regression, C-index, and AUC were used to assess model performance; SHAP analysis explored feature importance.

Main Results:

• The combined model demonstrated superior predictive performance compared to the clinical model alone (C-index 0.756 vs 0.713).
• Independent predictors included age, serum creatinine, segment stenosis score (SSS), and left anterior descending (LAD)-fat attenuation index (FAI).
• LAD-FAI emerged as the most significant prognostic factor for MACE post-TAVR.

Conclusions:

• A combined model integrating CCTA features and clinical data enhances the prediction of adverse outcomes after TAVR.
• This approach offers improved prognostic accuracy for TAVR patients.
• CCTA-derived imaging biomarkers, particularly LAD-FAI, are valuable in risk stratification for TAVR.