Redo Surgical Aortic Valve Replacement After Prior Transcatheter Versus Surgical Aortic Valve Replacement.

BACKGROUND: Aortic stenosis treatment should consider risks and benefits for lifetime management. Although the feasibility of redo transcatheter aortic valve replacement (TAVR) remains unclear, concerns are emerging regarding reoperation after TAVR. OBJECTIVES: The authors sought to define comparative risk of surgical aortic valve replacement (SAVR) after prior TAVR or SAVR. METHODS: Data on patients undergoing bioprosthetic SAVR after TAVR and/or SAVR were extracted from the Society of Thoracic Surgeons Database (2011-2021). Overall and isolated SAVR cohorts were analyzed. The primary outcome was operative mortality. Risk adjustment using hierarchical logistic regression as well as propensity score matching for isolated SAVR cases were performed. RESULTS: Of 31,106 SAVR patients, 1,126 had prior TAVR (TAVR-SAVR), 674 had prior SAVR and TAVR (SAVR-TAVR-SAVR), and 29,306 had prior SAVR (SAVR-SAVR). Yearly rates of TAVR-SAVR and SAVR-TAVR-SAVR increased over time, whereas SAVR-SAVR was stable. The TAVR-SAVR patients were older, with higher acuity, and with greater comorbidities than other cohorts. The unadjusted operative mortality was highest in the TAVR-SAVR group (17% vs 12% vs 9%, respectively; P < 0.001). Compared with SAVR-SAVR, risk-adjusted operative mortality was significantly higher for TAVR-SAVR (OR: 1.53; P = 0.004), but not SAVR-TAVR-SAVR (OR: 1.02; P = 0.927). After propensity score matching, operative mortality of isolated SAVR was 1.74 times higher for TAVR-SAVR than SAVR-SAVR patients (P = 0.020). CONCLUSIONS: The number of post-TAVR reoperations is increasing and represent a high-risk population. Yet even in isolated SAVR cases, SAVR after TAVR is independently associated with increased risk of mortality. Patients with life expectancy beyond a TAVR valve and unsuitable anatomy for redo-TAVR should consider a SAVR-first approach.

Interpretable machine learning-based predictive modeling of patient outcomes following cardiac surgery.

BACKGROUND: The clinical applicability of machine learning predictions of patient outcomes following cardiac surgery remains unclear. We applied machine learning to predict patient outcomes associated with high morbidity and mortality after cardiac surgery and identified the importance of variables to the derived model's performance. METHODS: We applied machine learning to the Society of Thoracic Surgeons Adult Cardiac Surgery Database to predict postoperative hemorrhage requiring reoperation, venous thromboembolism (VTE), and stroke. We used permutation feature importance to identify variables important to model performance and a misclassification analysis to study the limitations of the model. RESULTS: The study dataset included 662,772 subjects who underwent cardiac surgery between 2015 and 2017 and 240 variables. Hemorrhage requiring reoperation, VTE, and stroke occurred in 2.9%, 1.2%, and 2.0% of subjects, respectively. The model performed remarkably well at predicting all 3 complications (area under the receiver operating characteristic curve, 0.92-0.97). Preoperative and intraoperative variables were not important to model performance; instead, performance for the prediction of all 3 outcomes was driven primarily by several postoperative variables, including known risk factors for the complications, such as mechanical ventilation and new onset of postoperative arrhythmias. Many of the postoperative variables important to model performance also increased the risk of subject misclassification, indicating internal validity. CONCLUSIONS: A machine learning model accurately and reliably predicts patient outcomes following cardiac surgery. Postoperative, as opposed to preoperative or intraoperative variables, are important to model performance. Interventions targeting this period, including minimizing the duration of mechanical ventilation and early treatment of new-onset postoperative arrhythmias, may help lower the risk of these complications.