Traumatic and hemorrhagic complications after extracorporeal cardiopulmonary resuscitation for out-of-hospital cardiac arrest

Publication Date


Journal Title



© 2020 Elsevier B.V. Introduction: Extracorporeal cardiopulmonary resuscitation (ECPR) is an emerging invasive rescue therapy for treatment of refractory out-of-hospital cardiac arrests (OHCA). We aim to describe the incidence of traumatic and hemorrhagic complications among patients undergoing ECPR for OHCA and examine the association between CPR duration and ECPR-related injuries or bleeding. Methods: We examined prospectively collected data from the Extracorporeal Resuscitation Outcomes Database (EROD), which includes ECPR-treated OHCAs from participating hospitals (October 2014 to August 2019). The primary outcome was traumatic or hemorrhagic complications, defined any of the following: pneumothorax, pulmonary hemorrhage, major bleeding, cannula site bleeding, gastrointestinal bleeding, thoracotomy, cardiac tamponade, aortic dissection, or vascular injury during hospitalization. The primary exposure was the cardiac arrest to ECPR initiation interval (CA-ECPR interval), measured as the time from arrest to initiation of ECPR. Descriptive statistics were used to compare demographic, cardiac arrest, and ECPR characteristics among patients with and without CPR-related traumatic or bleeding complications. Multivariable logistic regression was used to examine the association between CA-ECPR interval and traumatic or bleeding complications. Results: A total of 68 patients from 4 hospitals receiving ECPR for OHCA were entered into EROD and met inclusion criteria. Median age was 51 (interquartile range 38–58), 81% were male, 40% had body mass index > 30, and 70% had pre-existing medical comorbidities. A total of 65% had an initial shockable cardiac rhythm, mechanical CPR was utilized in at least 29% of patients, and 27% were discharged alive. The median time from arrest to ECPR initiation was 73 min (IQR 60–104). A total of 37% experienced a traumatic or bleeding complication, with major bleeding (32%), vascular injury (18%), and cannula site bleeding (15%) being the most common. Compared to patients with shorter CPR times, patients with a longer CA-ECPR interval had 18% (95% confidence interval — 2–42%) higher odds of suffering a mechanical or bleeding complication, but this did not reach statistical significance (p = 0.08). Conclusions: Traumatic injuries and bleeding complications are common among patients undergoing ECPR. Further study is needed to investigate the relation between arrest duration and complications. Clinicians performing ECPR should anticipate and assess for injuries and bleeding in this high-risk population.

Document Type





School of Medicine

Primary Department

Emergency Medicine





For the public and Northwell Health campuses