1Department of General Surgery, Meenakshi Medical College Hospital & Research Institute, Meenakshi Academy of Higher Education and Research
2Department of Cardiology, Meenakshi College of Allied Health Sciences & Meenakshi Medical College Hospital & Research Institute, Meenakshi Academy of Higher Education and Research
3Professor, Meenakshi College of Pharmacy, Meenakshi Academy of Higher Education and Research
4Department of Pharmacology, Meenakshi Ammal Dental College and Hospital, Meenakshi Academy of Higher Education and Research
5Arulmigu Meenakshi College of Nursing, Meenakshi Academy of Higher Education and Research
6Meenakshi College of Physiotherapy, Meenakshi Academy of Higher Education and Research
Background: Transcatheter Aortic Valve Replacement (TAVR) now can be considered as a preferred intervention in patients with severe aortic stenosis however the major concerns are with regards to peri-operative complications. Older risk-assessment tools are not formally designed to identify these kinds of patterns of subtle physiological patterns that require sophisticated techniques of data analysis.
Objective: The proposed study aims to assess the performance of a risk-stratification model based on the AI-driven preoperative risk review in predicting both procedural and 30-day adverse events in patients undergoing TAVR therapies in different centres.
Method: It was an observational multicentre study focusing on patients who underwent TAVR in 2021-2024 in the five tertiary hospitals. The combination of pre-operative clinical and imaging and biomarker data was done with a machine-learning model under supervision. The model performance according to the cross-validation was determined and compared to the common clinical risk scores. Its main outcomes were the major adverse cardiovascular events (MACE) and 30-day unplanned readmission.
Results: The prediction accuracy by AI model was superior to the traditional scores that had AUC of 0.89 compared to 0.74 with the conventional tools. Dependent variables were the measures of ventricular strain, frailty indices and biomarkers of inflammation. Patients who were at a high risk identified by the algorithm had much higher post-procedural complication rates and early readmission.
Conclusion: The AI-based pre-operative risk stratification has a high promise of beneficial effects on patient selection and peri-procedural strategies in TAVR. Its successful adoption could speed up individualised decision-making and minimise initially negative results.
Keywords: Artificial Intelligence, risk stratification, predictive modeling, frailty measures, cardiac imaging, risk prediction models.
How to cite this article: Aravind P, Jayakodi T, Chamundeeswari D, Packirisamy S, Kalpana P, Divya S. Pre-operative AI-Driven Risk Stratification for Transcatheter Aortic Valve Replacement: A Multicentre Study. Int J Drug Deliv Technol. 2026;16(10s): 202-208; DOI: 10.25258/ijddt.16.10s.30
Source of support: Nil.
Conflict of interest: None