Estimating Mortality Risk for Adult Congenital Heart Surgery: An Analysis of The Society of Thoracic Surgeons Congenital Heart Surgery Database

Abstract

Adjustment for case mix is critical to accurate outcomes analysis in congenital heart surgery. Established tools encompass all age groups and are not specific to the growing population of adults undergoing congenital heart operations. We derived an empirically based adult congenital heart surgery (ACHS) mortality score. In-hospital mortality was analyzed for the 152 most common procedures/procedural groups in adults 18 years of age and older in The Society of Thoracic Surgeons Congenital Heart Surgery Database (STS-CHSD) (2000-2013). Procedure-specific adult mortality rate estimates were calculated using Bayesian methods adjusting for small denominators for procedures with 30 cases or more (N = 52). Each procedural group was assigned an ACHS mortality score ranging from 0.1 to 3.0 based on the estimated mortality rate. Discrimination was assessed using the c-index in a separate validation sample. A total of 12,513 procedures (116 centers) were analyzed. Overall unadjusted mortality was 1.8%. Significant differences in mortality rates in adults compared with all ages were seen for several procedures, including Ebstein's repair (0.7% versus 4.9%; p = 0.003) and Fontan operations (6.8% versus 1.4%; p < 0.01). The procedure with the lowest model-based estimate of mortality and accompanying ACHS mortality score was atrial septal defect repair (0.2%, 0.1), and the highest was Fontan revision (9.7%, 3.0). The c-index for the ACHS mortality score was 0.809 versus 0.777 for the "non-age-specific" Society of Thoracic Surgeons-European Association for Cardio-thoracic Surgery (STAT) mortality score applied to adults. Risk estimation based on the aggregate of all age groups is suboptimal when analyzing outcomes specifically among adults. An empirically based ACHS mortality score can facilitate case-mix adjustment by providing accurate estimation of mortality risk for adults.