Strategies for Anticoagulation During Venovenous ECMO

Since the inception of Extracorporeal Membrane Oxygenation (ECMO), moderate intensity titrated dose anticoagulation has been used to prevent clinically harmful thromboembolism and thrombotic mechanical complications. The impact of thromboembolic events on clinical outcomes during venovenous (V-V) extracorporeal membrane oxygenation (ECMO), however, is unclear, and complications related to bleeding are common and associated with increased morbidity and mortality. These findings have led many experts to suggest that anticoagulation strategies during V-V ECMO should be re-evaluated.

Critical illness, in general, is associated with both coagulopathy and impaired hemostasis. These problems are compounded during ECMO by the artificial interface between blood and the non-biologic surface of the circuit components, which leads to activation of the coagulation system, consumptive thrombocytopenia, fibrinolysis, and thrombin generation. The sheer stress on blood components during ECMO also lead to destruction of high-molecular-weight von Willebrand multimers, interrupting primary hemostasis.

Both bleeding and thromboembolism are common complications during ECMO. Bleeding events have been associated with poor clinical outcomes, likely mediated by an increased incidence of intracranial hemorrhage during ECMO. During intra-operative cardiopulmonary bypass and venoarterial (V-A) ECMO, ischemic strokes are a common and potentially deadly complication. During V-V ECMO, however, the majority of thromboembolic events are cannula-associated DVT and circuit thromboses requiring exchange, which are of unclear clinical significance.

Various anticoagulation strategies have been proposed to balance the risks of bleeding and thromboembolism during V-V ECMO, including high intensity anticoagulation, moderate intensity anticoagulation, and low intensity anticoagulation (the equivalent of DVT prophylaxis). Observational studies have suggested that, compared to moderate intensity anticoagulation, low intensity anticoagulation reduces transfusion requirements without affecting the incidence of thrombosis, hemorrhage, or death. In one case series of 60 patients who were treated with only low-intensity subcutaneous heparin during V-V ECMO, rates of transfusions were lower than historical controls without any effect on the rate of thrombotic events. Similarly, a recent systematic review suggested that the rates of thromboembolism and circuit thrombosis among patients managed with a moderate intensity anticoagulation strategy during V-V ECMO were comparable to the rates reported among patients managed with a less intense anticoagulation strategy.

To date, there are no randomized controlled trials comparing low intensity to moderate intensity anticoagulation during V-V ECMO. Guidelines from the Extracorporeal Life Support Organization (ELSO), the pre-eminent group for ECMO education and research, provide little guidance for the selection of anticoagulation strategy, and anticoagulation practices are highly variable across institutions. A large, multicenter, randomized trial is needed to determine the ideal strategy to anticoagulation during V-V ECMO. Before such a trial can be conducted, however, additional data are needed on the feasibility of randomizing patients to a specific anticoagulation strategy and study measurements.

To facilitate a large, multicenter randomized controlled trial comparing low intensity anticoagulation to moderate intensity anticoagulation during V-V ECMO, a pilot trial is needed to establish feasibility and the performance of the primary outcome measures.

Primary aim of the study: To demonstrate feasibility of a future large, multi-center randomized controlled trial comparing low intensity to moderate intensity anticoagulation among adults receiving V-V ECMO by demonstrating the ability to recruit and randomize participants, adhere to assigned anticoagulation strategy, and demonstrate adequate separation between groups in therapy delivered and intensity of anticoagulation achieved with the assigned anticoagulation strategies.

Secondary aim of the study: To define and estimate the frequency of the primary efficacy, primary safety, and secondary outcomes of a future large, multi-center randomized controlled trial comparing low intensity vs moderate intensity anticoagulation among adults receiving V-V ECMO.

Inclusion criteria

• Patient receiving V-V ECMO
• Patient is located in a participating unit of the Vanderbilt University Medical Center (VUMC) adult hospital.

Exclusion criteria

• Patient is pregnant
• Patient is a prisoner
• Patient is < 18 years old
• Patient underwent ECMO cannulation greater than 24 hours prior to screening
• Presence of an indication for systemic anticoagulation:

• Ongoing receipt of systemic anticoagulation
• Planned administration of anticoagulation for an indication other than ECMO
• Presence of or plan to insert an arterial ECMO cannula
• Presence of a contraindication to anticoagulation:

• Active bleeding determined by treating clinicians to make anticoagulation unsafe
• Major surgery or trauma less than 72 hours prior to randomization
• Known history of a bleeding diathesis
• Ongoing severe thrombocytopenia (platelet count < 30,000)
• History of heparin-induced thrombocytopenia (HIT)
• Heparin allergy
• Positive SARS-CoV-2 test within prior 21 days or high clinical suspicion for COVID-19
• The treating clinician determines that the patient's risks of thromboembolism or bleeding necessitate a specific approach to anticoagulation management during V-V ECMO