Long-Term Outcomes of Out-of-Hospital Cardiac Arrest Care at Regionalized Centers

Jonathan Elmer, MD, MS
elmerjp@upmc.edu
Clifton W. Callaway, MD, PhD, Chung-Chou H. Chang, PhD, Jonathan Madaras, MPA, NRP, Christian Martin-Gill, MD, MPH, Philip Nawrocki, MD, Kristen A.C. Seaman, MD, Denisse Sequeira, BS, Owen T. Traynor, MD, Arvind Venkat, MD, Heather Walker, MD, David J. Wallace, MD, MPH, Francis X. Guyette, MD, MS
annemergmed.2018.05.018

Study objective
It is unknown whether regionalization of postarrest care by interfacility transfer to cardiac arrest receiving centers reduces mortality. We seek to evaluate whether treatment at a cardiac arrest receiving center, whether by direct transport or early interfacility transfer, is independently associated with long-term outcome.

Methods
We performed a retrospective cohort study including adults resuscitated from out-of-hospital cardiac arrest in southwestern Pennsylvania and neighboring Ohio, West Virginia, and Maryland, which includes approximately 5.7 million residents in urban, suburban, and rural counties. Patients were treated by 1 of 78 ground emergency medical services agencies or 2 air medical transport agencies between January 1, 2010, and November 30, 2014. Our primary exposures of interest were interfacility transfer to a cardiac arrest receiving center within 24 hours of arrest or any treatment at a cardiac arrest receiving center regardless of transfer status. Our primary outcome was vital status, assessed through December 31, 2014, with National Death Index records. We used unadjusted and adjusted survival analyses to test the independent association of cardiac arrest receiving center care, whether through direct or interfacility transport, on mortality.

Results
Overall, 5,217 cases were observed for 3,629 person-years, with 3,865 total deaths. Most patients (82%) were treated at 42 non–cardiac arrest receiving centers with median annual volume of 17 cases (interquartile range 1 to 53 cases per center annually), whereas 18% were cared for at cardiac arrest receiving centers receiving at least 1 interfacility transfer per month. In adjusted models, treatment at a cardiac arrest receiving center was independently associated with reduced hazard of death compared with treatment at a non–cardiac arrest receiving center (adjusted hazard ratio 0.84; 95% confidence interval 0.74 to 0.94). These effects were unchanged when analysis was restricted to patients brought from the scene to the treating hospital. No other hospital characteristic, including total out-of-hospital cardiac arrest patient volume and cardiac catheterization capabilities, independently predicted outcome.

Conclusion
Both early interfacility transfer to a cardiac arrest receiving center and direct transport to a cardiac arrest receiving center from the scene are independently associated with reduced mortality.