2025 Southeastern Residency Conference

Title: 
Efficacy and Safety of Bivalirudin Compared with Heparin Anticoagulation in Critically Ill Patients Requiring Venovenous Extracorporeal Membrane Oxygenation


Authors: 
Brianna R. Landrum; Nick Tran; Kayla A. Lawlor; Sagar B. Dave; Christina Creel-Bulos; Casey Miller; Jolie Gallagher


Objective: 
To evaluate the efficacy and safety of bivalirudin compared with heparin anticoagulation in critically ill patients requiring VV-ECMO.


Self Assessment Question: 
The ELSO General Guidelines currently recommend unfractionated heparin as the preferred
anticoagulant for patients requiring VV-ECMO.
a) True
b) False


Background: 
Extracorporeal membrane oxygenation (ECMO) is an invasive mechanical circulatory support
system that is utilized in the management of critically ill patients. Veno-venous ECMO (VV-ECMO) is primarily used for patients with respiratory failure, and anticoagulation is crucial to prevent thrombosis. Both unfractionated heparin (UFH) and bivalirudin are commonly used anticoagulants in ECMO; however, the data comparing these agents in VV-ECMO is sparse. Therefore, this study aims to evaluate the efficacy and safety of bivalirudin compared with heparin anticoagulation in critically ill patients requiring VV-ECMO.


Methods:
This retrospective study was conducted at Emory University Hospital (EUH), including adult
patients (≥18 years) who received VV-ECMO between January 2020 and April 2024. Patients who were anticoagulated with either UFH or bivalirudin within 24 hours of VV-ECMO cannulation were included. Exclusion criteria included pregnancy, history of factor deficiencies, history of Chronic Thromboembolic Pulmonary Hypertension (CTEPH), VA-ECMO cannulation, single-site cannulation, death or withdrawal of care within 24 hours of VV-ECMO cannulation, or ECMO managed at an outside hospital (OSH) for greater than 30 days. The primary outcome was the incidence of VV-ECMO circuit or oxygenator exchange. Secondary outcomes included time spent within the therapeutic anti-Xa and activated partial thromboplastin time (aPTT) ranges, ICU and hospital length of stay, duration of mechanical ventilation and VV-ECMO support, and the incidence of systemic thrombosis and post-decannulation thrombosis. Safety outcomes included the total volume of blood products transfused, the incidence of new hemorrhage, the occurrence of thrombocytopenia or heparin-induced thrombocytopenia (HIT), and mortality rates, including both inpatient and pre-decannulation mortality, as well as mortality at 28 days.


Results:
A total of 124 patients were included, with 75 receiving UFH and 49 receiving bivalirudin.
Baseline characteristics were similar between the two groups, though patients receiving bivalirudin had a higher Sequential Organ Failure Assessment (SOFA) score (10 for UFH vs. 12 for bivalirudin, p=0.007) and a larger number of patients with COVID-19 as a primary diagnosis [3 (0.04%) for UFH vs. 26 (53.06%) for bivalirudin]. There were no significant differences in the primary outcome of circuit or oxygenator exchange between the two groups (8.0% for UFH vs. 12.2% for bivalirudin, p=0.538). However, patients on bivalirudin spent a significantly greater proportion of time within the therapeutic aPTT range compared to UFH (58.5% vs. 33.3%, p<0.001). Secondary outcomes showed that patients in the bivalirudin group had significantly longer ICU and inpatient stays, as well as longer durations of mechanical ventilation and VV-ECMO. Both groups had similar rates of hemorrhage and thrombosis, but the bivalirudin group had a significantly higher rate of in-hospital (40.8% vs. 17.3%, p=0.004) and pre-decannulation mortality (36.7% vs. 16.0%, p=0.008).


Conclusion:
In patients receiving VV-ECMO, there was no difference in the rate of circuit or oxygenator
exchange between bivalirudin or UFH. Bivalirudin was associated with longer ICU and hospital length of stay, increased mechanical ventilation and ECMO duration, and higher in-hospital and pre-decannulation mortality compared to UFH. However, there were no differences in the rate of thrombosis or hemorrhage between the two anticoagulants. The observed differences in outcomes may be attributed to the higher severity of illness in the bivalirudin group, particularly among patients with COVID-19. Despite bivalirudin providing more predictable anticoagulation, these findings suggest that further research with larger, prospective studies and standardized anticoagulation protocols is needed to better define the optimal anticoagulant for VV-ECMO patients.