Development and internal validation of risk scores to predict survival in the pediatric population following in-hospital cardiac arrest.

IntroductionIn-hospital cardiac arrest (IHCA) in the pediatric population is associated with poor survival and neurological outcomes. We aimed to develop and internally validate a risk score to predict survival to discharge following pediatric IHCA. MethodsWe included pediatric IHCA patients in the Get With The Guidelines-Resuscitation(R) registry between 2005 and 2021. We used logistic regression (LR), classification and regression trees (CART), and artificial neural networks (ANN) to develop models using 70% of the data and validate them using the remaining 30% of the data. Discrimination was based on the area under the receiver operating characteristic curve (AUC), and predictive accuracy on percent survival in each risk group. ResultsWe included 6141 patients with a mean age of 4.8 years, of whom 41.3% (n = 2535) were infants < 1 year of age and 39.1% of whom survived to hospital discharge. We developed separate models for infants and older children. The most important independent pre-arrest predictors were age, illness category, acyanotic cardiac malformation, cyanotic cardiac malformation, hepatic insufficiency, hypotension/hypoperfusion, metabolic/electrolyte abnormality, metastatic/hematologic malignancy, renal insufficiency, congenital malformation, septicemia, hypotension, trauma, and pediatric cerebral performance score on admission. All three approaches showed good classification accuracy in the derivation sample (AUC for LR: 0.70, 0.71, AUC for CART: 0.68, 0.70, AUC for ANN: 0.76, 0.74 for infants and older children respectively) and used almost the same number of variables. Logistic regression and CART models were the most useful as they identified patients with the lowest survival, showed good discrimination, and could be used to develop a simple point score and decision trees that can be implemented in the clinical or research setting. In infants, the average probability of survival was 10%, 36%, and 60% whereas in older children it was 6.2%, 31.1%, and 62.3% in the low, moderate, and high survival categories in the LR model. ConclusionPediatric patients experiencing IHCA can be classified into low, moderate, and high survival categories using a simple risk score and easily identified pre-arrest variables. These risk scores can support clinical decisions, facilitate research, and help monitor the quality of medical services.