Abstract 4138620: Assessment of the Accuracy of the Perioperative Mortality Risk in Adults with Congenital Heart Disease (PEACH) Score

INTRODUCTION: Previous mortality risk scores for adult cardiac surgery patients did not include adult congenital heart disease (ACHD) patients. Development of ACHD specific mortality risk scores include the procedure-specific ACHS score and recently derived PEACH score. While validated in Europe, further evaluation of the PEACH score in additional settings is necessary. RESEARCH QUESTION: We assessed the efficacy of the PEACH score to predict mortality in ACHD patients undergoing congenital cardiac surgery at our institution. We compared the PEACH score to the ACHS and STS scores and assessed for additional risk factors which may further augment the PEACH score. Methods: A retrospective cohort study was conducted of patients who underwent congenital cardiac surgery at the University of Colorado Hospital and Children’s Hospital Colorado between 2010-2022. Patients were identified by the institutional STS congenital cardiac surgery database and were eligible if 18 years or older and underwent congenital cardiac surgery. Components of the PEACH score and additional postulated risk factors were evaluated by univariate analysis. A receiver operating characteristic (ROC) curve was used to compare the PEACH score to existing risk models. Logistic regression analysis determined additional risk factors to augment the PEACH score. Results: A total of 516 patients were included with a median age of 30 (IQR 23-42) and 51% were male. There were nine deaths overall. The PEACH score was predictive of mortality (OR: 2.09, p<0.001) with a score of 0: 0 deaths/240 patients (0%); 1-2: 6 deaths/229 patients (2.6%); 3+: 3 deaths/41 patients (7.3%). Of additional risk factors tested, platelets less than 184,000 per mcL (OR: 8.36, p=0.01) was significantly associated with mortality. Comparison of the models revealed PEACH score AUC 0.832 (95% CI: 0.740,0.924), ACHS score AUC 0.869 (95% CI: 0.806, 0.932), and STAT AUC 0.769 (95% CI: 0.616, 0.922, Figure 1). The addition of platelet level improved the predictive ability of the PEACH score (AUC 0.881). Conclusion: In this single-center analysis, the PEACH score provided strong predictive ability for peri-operative mortality. The addition of platelet levels may further augment this score.