Machine Learning-Based Mortality Prediction of Patients Undergoing Type A Aortic Dissection Surgery

Background: The lack of appropriate, well-validated, and specific means to risk-stratify patients with acute type A aortic dissection (ATAAD) complicates the evaluation of perioperative survival in clinical practice. We aim to develop an inflammation-based risk stratification tool for operative mortality in ATAAD patients using machine learning. Methods: Between 2016 and 2021, 3124 patients from Beijing Anzhen Hospital were included for derivation, 571 patients from same hospital for internal validation; 1319 patients form other 12 hospitals for external validation. Primary outcome was operative mortality according to STS criteria. LASSO regression was used to identify clinical risk factors. Model was developed using different ML algorithms. Model performances was determined using area under the receiver operating characteristic curve (AUC) for discrimination, calibration curves, and Brier score for calibration. The final model, called 5A score was tested with respect to the existing clinical scores. Findings: Extreme gradient boosting was selected for model training (5A Score), using 12 variables for prediction: the ratio of platelet to leukocyte count, creatinine, age, hemoglobin, prior cardiac surgery, extent of dissection extension, cerebral perfusion, aortic regurgitation, sex, pericardial effusion, shock, and coronary perfusion, which yields the highest AUC (0.873 [0.845-0.901]). 5A score’s AUC was 0.875 (0.814-0.936), 0.845 (0.811-0.878) and 0.852 (0.821-0.883) in the internal, external and total cohort, which outperformed the best existing risk score (GERAADA Score AUC 0.709 [0.669-0.749]). Interpretation: The 5A Score is a novel, internally and externally validated inflammation-based tool for risk stratification of patients prior to surgical repair, potentially advancing individual anti-inflammatory therapy.Trial Registration Details: This study was registered with Clinical Trials. gov number NCT04918108. Funding Information: The National Natural Science Foundation of China (82000305) and Scientific Research Common Program of Beijing Municipal Commission of Education (KM202110025014). Declaration of Interests: We declare no competing interests.Ethics Approval Statement: The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki, and was approved by the Institutional Review Board of Aortic Collaborative Institutions involved (2021-SR-381). Informed consent was waived for this retrospectively observational study.