What is the clinical evidence from this study?

Study design: Meta-Analysis. Population: severe aortic stenosis undergoing transcatheter aortic valve replacement (TAVR) (n=366269). Intervention: Artificial intelligence (AI) models. Primary outcome: All-cause mortality prediction (AUC 0.78, 95% CI 0.74-0.82).

July 11, 2025Open Access

Artificial Intelligence in Risk Stratification and Outcome Prediction for Transcatheter Aortic Valve Replacement: A Systematic Review and Meta-Analysis

Q: What does this research mean for the field?

Artificial intelligence models demonstrate promising performance in predicting post-TAVR all-cause mortality (AUC 0.78) and other complications, particularly when incorporating baseline clinical, imaging, and biomarker data. Novelty: ClaimNovelty.SYNTHESIS. Consensus alignment: ConsensusAlignment.NEUTRAL.

Key Result

Artificial intelligence models demonstrated promising performance in predicting post-TAVR all-cause mortality (AUC 0.78; 95% CI 0.74-0.82).

Study Design

Type

Meta-Analysis (n=366,269)

Structured PICO

Do artificial intelligence models accurately predict post-TAVR outcomes and complications in patients undergoing transcatheter aortic valve replacement?

Population

366,269 patients following transcatheter aortic valve replacement (TAVR) pooled from 43 studies, mean age 80 ± 8.25, 52.9% men.

Intervention

Artificial intelligence (AI) and machine learning algorithms (including random forests, neural networks, extreme gradient boosting, and support vector machines) for risk stratification and outcome prediction.

Outcome

Model performance metrics (recall, area under the curve [AUC], and accuracy) for predicting post-TAVR outcomes including all-cause mortality, permanent pacemaker implantation or new left bundle branch block, valve-related dysfunction, and major adverse cardiovascular events.surrogate

AI-based risk prediction models show promising performance for predicting post-TAVR complications, particularly when incorporating baseline clinical, imaging, and biomarker data.

Main Result

Effect estimate: AUC 0.78 (95% CI 0.74-0.82)

Limitations

Requires evaluation in external validation datasets

Abstract

Background/Objectives: Transcatheter aortic valve replacement (TAVR) has been introduced as an optimal treatment for patients with severe aortic stenosis, offering a minimally invasive alternative to surgical aortic valve replacement. Predicting these outcomes following TAVR is crucial. Artificial intelligence (AI) has emerged as a promising tool for improving post-TAVR outcome prediction. In this systematic review and meta-analysis, we aim to summarize the current evidence on utilizing AI in predicting post-TAVR outcomes. Methods: A comprehensive search was conducted to evaluate the studies focused on TAVR that applied AI methods for risk stratification. We assessed various ML algorithms, including random forests, neural networks, extreme gradient boosting, and support vector machines. Model performance metrics—recall, area under the curve (AUC), and accuracy—were collected with 95% confidence intervals (CIs). A random-effects meta-analysis was conducted to pool effect estimates. Results: We included 43 studies evaluating 366,269 patients (mean age 80 ± 8.25; 52.9% men) following TAVR. Meta-analyses for AI model performances demonstrated the following results: all-cause mortality (AUC = 0.78 (0.74–0.82), accuracy = 0.81 (0.69–0.89), and recall = 0.90 (0.70–0.97); permanent pacemaker implantation or new left bundle branch block (AUC = 0.75 (0.68–0.82), accuracy = 0.73 (0.59–0.84), and recall = 0.87 (0.50–0.98)); valve-related dysfunction (AUC = 0.73 (0.62–0.84), accuracy = 0.79 (0.57–0.91), and recall = 0.54 (0.26–0.80)); and major adverse cardiovascular events (AUC = 0.79 (0.67–0.92)). Subgroup analyses based on the model development approaches indicated that models incorporating baseline clinical data, imaging, and biomarker information enhanced predictive performance. Conclusions: AI-based risk prediction for TAVR complications has demonstrated promising performance. However, it is necessary to evaluate the efficiency of the aforementioned models in external validation datasets.

Mark Helpful

Bookmark

Relay

View Full Paper