Early Risk Stratification of Patients After Resuscitation from Out-of-hospital Cardiac Arrest

Introduction: Over 400,000 people suffer an out-of-hospital cardiac arrest (OHCA) each year in Canada and the United States of which less than 10% survive to hospital discharge. Cardiac arrest is a heterogenous condition and patient outcomes are impacted by a multitude of factors. Post-cardiac arrest prognostication is recommended to occur 72 hours after return of spontaneous circulation, however it is not known if there are factors that can be utilized early in the post-cardiac arrest period to predict patient outcome. The objective of our study was to develop a novel clinical prediction rule that could be used to risk stratify patients early in the post-cardiac arrest period. Methods: We conducted a systematic review and meta-analysis to identify predictors of survival to include in our risk stratification model. We then performed a retrospective cohort study to derive and internally validate a number of statistical models to risk stratify cardiac arrest patients. We included data from 2010 to 2015 from the Epistry Cardiac Arrest database in Toronto and included both pediatric patients (cohort 1) and adult patients (cohort 2). Within each cohort we developed risk stratification models for all paramedic-treated OHCA and for patients who had a sustained in-hospital ROSC (>20 minutes). Using the results of our systematic review, as well as clinical expertise, we included a number of known predictors of patient outcome prediction and used ordinal and logistic regression analysis to derive our models. Models were internally validated using bootstrap validation. Results: We included a total of 18,527 adult and 379 pediatric patients from the Toronto ROC-Epistry Cardiac Arrest database between 2010 and 2015. In adult patients we were able to derive a clinical prediction model using Ustein variables, GCS motor score, duration of resuscitation, and pupillary response that predicted good neurological outcome with an AUC of 0.89 after internal validation. For pediatric patients we were able to predict good survival with an AUC of 0.81 using Utstein variables and duration of resuscitation. Conclusion: We were able to derive and internally validate a model to provide early risk stratification to cardiac arrest patients early in the post-cardiac arrest period. Future steps are to externally validate this model in another healthcare setting.Ph.D