Shahian, David M.

Background: Robust models exist to predict risk-adjusted mortality following congenital heart surgery. It is known that underlying diagnosis modifies predicted risk of operations, and institution-level outcomes vary based on their unique case mix. We therefore evaluated risk-adjusted outcomes for a subset of diagnosis-procedure (D-P) combinations within the Society of Thoracic Surgeons' (STS) congenital heart surgery database (CHSD). Methods: A clinician panel identified the most frequently encountered D-P combinations in CHSD between 07/2017 and 06/2021. Candidate modeling techniques were compared in a random 80% development cohort. Model performance was assessed using internal and cross-validation calibration, discrimination plots and optimism adjusted C-statistics. The final selected model was evaluated in the 20% validation sample. Results: The chosen subset of D-P combinations represented 45,384 of 87,589 (52%) unique episodes of surgical care in CHSD during this time-period. The development cohort (36,350 episodes) had 1.8% mortality (667 deaths). All candidate models had excellent discrimination with C-statistics ranging from 0.862 to 0.876. The model with the best combination of discrimination, calibration, and clinical face validity was a modified version of the current STS mortality risk model. This version included all current risk model variables with the addition of indicator variables for D-P combinations. In the 20% validation sample (9,034 episodes), this model was well calibrated (Figure) and had excellent discrimination (C = 0.879). Conclusion: We describe a cohort of D-P combinations widely encountered in the STS CHSD. Further, we refined the existing STS CHSD risk model for application to these D-P dyads to derive empirical risk-adjusted benchmark outcomes. This approach will benefit quality improvement efforts at an institutional level and may have implications for publicly reporting congenital heart surgery outcomes.

Background This study evaluated the role of supplementing Society of Thoracic Surgeons (STS) risk models for surgical aortic valve replacement with machine learning (ML). Methods and Results Adults undergoing isolated surgical aortic valve replacement in the STS National Database between 2007 and 2017 were included. ML models for operative mortality and major morbidity were previously developed using extreme gradient boosting. Concordance and discordance in predicted risk between ML and STS models were defined using equal‐size tertile‐based thresholds of risk. Calibration metrics and discriminatory capability were compared between concordant and discordant patients. A total of 243 142 patients were included. Nearly all calibration metrics were improved in cases of concordance. Similarly, concordance indices improved substantially in cases of concordance for all models with the exception of deep sternal wound infection. The greatest improvements in concordant versus discordant cases were in renal failure: ML model (concordance index, 0.660 [95% CI, 0.632–0.687] discordant versus 0.808 [95% CI, 0.794–0.822] concordant) and STS model (concordance index, 0.573 [95% CI, 0.549–0.576] discordant versus 0.797 [95% CI, 0.782–0.811] concordant) (each P <0.001). Excluding deep sternal wound infection, the concordance indices ranged from 0.549 to 0.660 for discordant cases and 0.674 to 0.808 for concordant cases. Conclusions Supplementing ML models with existing STS models for surgical aortic valve replacement may have an important role in risk prediction and should be explored further. In particular, for the roughly 25% to 50% of patients demonstrating discordance in estimated risk between ML and STS, there appears to be a substantial decline in predictive performance suggesting vulnerability of the existing models in these patient subsets.

Objectives: STAT Mortality Categories (developed 2009) stratify congenital heart surgery procedures into groups of increasing mortality risk to characterize case mix of congenital heart surgery providers. This update of the STAT Mortality Score and Categories is empirically based for all procedures and reflects contemporary outcomes. Methods: Cardiovascular surgical operations in the Society of Thoracic Surgeons Congenital Heart Surgery Database (January 1, 2010 – June 30, 2017) were analyzed. In this STAT 2020 Update of the STAT Mortality Score and Categories, the risk associated with a specific combination of procedures was estimated under the assumption that risk is determined by the highest risk individual component procedure. Operations composed of multiple component procedures were eligible for unique STAT Scores when the statistically estimated mortality risk differed from that of the highest risk component procedure. Bayesian modeling accounted for small denominators. Risk estimates were rescaled to STAT 2020 Scores between 0.1 and 5.0. STAT 2020 Category assignment was designed to minimize within-category variation and maximize between-category variation. Results: Among 161,351 operations at 110 centers (19,090 distinct procedure combinations), 235 types of single or multiple component operations received unique STAT 2020 Scores. Assignment to Categories resulted in the following distribution: STAT 2020 Category 1 includes 59 procedure codes with model-based estimated mortality 0.2% to 1.3%; Category 2 includes 73 procedure codes with mortality estimates 1.4% to 2.9%; Category 3 includes 46 procedure codes with mortality estimates 3.0% to 6.8%; Category 4 includes 37 procedure codes with mortality estimates 6.9% to 13.0%; and Category 5 includes 17 procedure codes with mortality estimates 13.5% to 38.7%. The number of procedure codes with empirically derived Scores has grown by 58% (235 in STAT 2020 vs 148 in STAT 2009). Of the 148 procedure codes with empirically derived Scores in 2009, approximately one-half have changed STAT Category relative to 2009 metrics. The New STAT 2020 Scores and Categories demonstrated good discrimination for predicting mortality in an independent validation sample (July 1, 2017-June 30, 2019; sample size 46,933 operations at 108 centers) with C-statistic = 0.791 for STAT 2020 Score and 0.779 for STAT 2020 Category. Conclusions: The updated STAT metrics reflect contemporary practice and outcomes. New empirically based STAT 2020 Scores and Category designations are assigned to a larger set of procedure codes, while accounting for risk associated with multiple component operations. Updating STAT metrics based on contemporary outcomes facilitates accurate assessment of case mix.

Background— The use of transcatheter aortic valve replacement (TAVR) to treat aortic stenosis in the United States is growing, yet little is known about the variation in procedural outcomes in community practice. We developed a TAVR in-hospital mortality risk model and used it to quantify variation in mortality rates across United States (US) TAVR centers. Methods and Results— We analyzed data from 22 248 TAVR procedures performed at 318 sites participating in the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry (November 2011 to October 2014). A Bayesian hierarchical model was developed to estimate hospital-specific risk-adjusted mortality rates adjusting for 40 patient baseline factors. A total of 1130 in-hospital deaths (5.1%) were observed. Reliability-adjusted risk-adjusted mortality rate estimates ranged from 3.4% to 7.7% with an interquartile range of 4.8% to 5.4%. A patient’s predicted odds of dying was 80% higher if treated by a hospital 1 standard deviation above the mean compared with a hospital 1 standard deviation below the mean (odds ratio =1.8; 95% credible interval, 1.4%–2.2%). Conclusions— Risk modeling of TAVR in-hospital mortality revealed variation in risk-adjusted mortality rates during the US early commercial experience. Transcatheter Valve Therapy Registry analyses using this model will support research, feedback reporting, and the identification of factors associated with quality.