Modeling Major Adverse Outcomes of Pediatric and Adult Patients With Congenital Heart Disease Undergoing Cardiac Catheterization

• DOI: 10.1161/circulationaha.117.027714
• Publication Date: 2017-09-08T00:50:16Z
• Authors: Natalie Jayaram; John A. Spertus; Kevin F. Kennedy; Robert Vincent; Gerard R. Martin; Jeptha P. Curtis; David Nykanen; Phillip M. Moore; Lisa Bergersen
• Keywords: Cardiac catheterization

Background: Risk standardization for adverse events after congenital cardiac catheterization is needed to equitably compare patient outcomes among different hospitals as a foundation quality improvement. The goal of this project was develop risk-standardization methodology adjust characteristics when comparing major in the NCDR’s (National Cardiovascular Data Registry) IMPACT Registry (Improving Pediatric and Adult Congenital Treatment). Methods: Between January 2011 March 2014, 39 725 consecutive patients within undergoing were identified. Given heterogeneity interventional procedures heart disease, new procedure-type risk categories derived with empirical data expert opinion, markers hemodynamic vulnerability. A multivariable hierarchical logistic regression model identify procedural predictive event or death 70% cohort validated remaining 30%. Results: rate 7.1% 7.2% derivation validation cohorts, respectively. Six 6 independent indicators vulnerability final adjustment included category, number indicators, renal insufficiency, single-ventricle physiology, coagulation disorder. had good discrimination, C-statistic 0.76 0.75 Model calibration excellent, slope 0.97 (standard error, 0.04; P value [for difference from 1] =0.53) an intercept 0.007 0.12; 0] =0.95). Conclusions: creation can support reporting risk-adjusted