Abstract 18554: Patient, Cardiac Arrest and Hospital Characteristics That Predict Receipt of Coronary Angiography in Out-of-Hospital Cardiac Arrest Patients

Abstract

Introduction: Previous studies have suggested an association between coronary angiography and improved outcomes amongst post cardiac arrest patients. Our objective was to measure the association between patient and hospital-level characteristics and receipt of coronary angiography to generate hypotheses to inform a definitive trial. Methods: This was a population-based retrospective cohort study of data from 28 hospitals in Southern Ontario between March 1, 2010 and December 31, 2014. We included consecutive adult patients with atraumatic, OHCA, who achieved return of spontaneous circulation, and were alive 6 hours after hospital arrival. Multi-level logistic regression was used to measure the relationship between patient and hospital-level covariates and receipt of coronary angiography, adjusted for clustering and potential confounders. Results: During the period of study, 2678 consecutive patients met the inclusion criteria; mean age 66(±16), 68.3% male, 45.9% shockable initial rhythm, 84.2% comatose at hospital admission. Overall, 32.4% received coronary angiography and 21.8% received percutaneous coronary intervention (PCI). Coronary angiography use varied from 12.7% to 63.6% across the sites. Factors significantly associated with receiving coronary angiography included ST-elevation (OR=23.31, CI95 17.64-30.80), being comatose at hospital arrival (OR=0.15, CI95 0.10-0.23), shockable initial cardiac rhythm (OR=4.87, CI95 3.70-6.41), bystander AED use (OR=2.05, CI95 1.21-3.47), EMS-witnessed arrest (OR=1.80, CI95 1.16-2.78), initiation of therapeutic hypothermia (OR=1.96, CI95 1.38-2.79), initial admission to a PCI centre (OR=3.20, CI95 1.78-5.76), male sex (OR=1.43, CI95 1.07-1.90) and age (OR=0.98, CI95 0.97-0.99). Conclusions: There is significant variability in receipt of coronary angiography after cardiac arrest. We identified several patient and hospital-level factors that contribute to this variability. Future work should determine which post arrest patients will benefit most from urgent angiography and develop and evaluate knowledge translation strategies to ensure consistent delivery of best practices.