Impact of anticoagulation strategy and agents on extracorporeal membrane oxygenation therapy.

Abstract

Extracorporeal membrane oxygenation mandates balancing the risk of thromboembolic complications with bleeding. We aimed to evaluate pragmatic anticoagulation regimens during extracorporeal membrane oxygenation and compare thromboembolic and bleeding outcomes. This retrospective, single-center study reviewed patients on venovenous or venoarterial extracorporeal membrane oxygenation for a minimum of 24 hours over a 5-year period. The primary outcome was composite thromboembolic events per day of extracorporeal membrane oxygenation. Secondary outcomes included composite bleeding complications, percent of measured activated partial thromboplastin times in goal range, and comparing events with therapeutic anticoagulation for the majority of the extracorporeal membrane oxygenation run (>50% of time on extracorporeal membrane oxygenation) versus non-therapeutic anticoagulation (therapeutic anticoagulation <50% of time). For the primary analysis, 100 patients received heparin, 10 received bivalirudin, and 43 were transitioned between heparin and bivalirudin. No significant differences were identified comparing the heparin group to the bivalirudin (RR = 0.427, p = 0.156) or transitioned group (RR = 1.274, p = 0.325). There were no differences in the rate of bleeding events when comparing the heparin group to the bivalirudin (RR = 0.626, p = 0.250) or transitioned group (RR = 0.742, p = 0.116). An increased number of adjustments to the anticoagulants was associated with a statistically higher rate of bleeding events per day (p = 0.006). There were no differences in thromboembolic or bleeding events when comparing different anticoagulant regimens. Adjustments to the anticoagulants are more likely to occur when bleeding is observed. Due to variability in anticoagulation, there is a need to standardize anticoagulation with extracorporeal membrane oxygenation.