Improving risk prediction after transcatheter aortic valve implantation: A comparison of machine learning with traditional methods.

Abstract

Background: Accurate mortality prediction following transcatheter aortic valve implantation (TAVI) is essential for mitigating risk, shared decision-making, and periprocedural planning. Surgical risk models have demonstrated modest discriminative value for patients undergoing TAVI and are typically poorly calibrated. Machine learning (ML) models offer an alternative risk stratification that may offer improved predictive accuracy. We performed a systematic review and meta-analysis comparing ML models with traditional risk scores for prediction of all-cause mortality after TAVI. Methods: PubMed, EMBASE, Web of Science and Cochrane databases were searched until 16 December 2023 for studies comparing ML models with traditional statistical methods for event prediction after TAVI. The primary outcome was comparative discrimination measured by C-statistics with 95% confidence intervals between ML models and traditional methods in estimating the risk of all-cause mortality at 30-days and 1-year. Results: Sixteen models were assessed across 9 observational studies (29608 patients). The summary C-statistic of the top-performing ML models for all-cause mortality was 0.82 (95% CI, 0.77–0.87), compared to traditional methods 0.65 (95% CI, 0.62–0.68). The difference in C-statistic between ML models and traditional methods was 0.17 (95% CI 0.15–0.19, p<0.00001). Of all included studies, 2 models were externally validated. Calibration was inconsistently reported. Conclusion: ML models outperformed traditional risk scores in the discrimination of all-cause mortality following TAVI. While integration of ML algorithms into electronic healthcare systems may improve periprocedural risk stratification, immediate implementation in the clinical setting remains uncertain. Further research is required to overcome methodological and validation limitations.