Extracorporeal Anticoagulation With Bivalirudin; Is the Best Still to Come?

To the Editor: We read with great interest the Brief Communication by Seelhammer et al. in response to our manuscript “Antithrombin Dose Optimization in Extracorporeal Membrane Oxygenation in Infants”.1 The authors write in support of direct Thrombin inhibitors (DTI) as alternatives to unfractionated heparin (UFH) for the systemic anticoagulation of patients supported by extracorporeal membrane oxygenation (ECMO). Because anticoagulation with UFH requires a multitude of laboratory tests and sufficient levels of the cofactor antithrombin for effect, they suggest the time may be right for DTIs to replace UFH as the first line anticoagulant for adult and pediatric patients supported on ECMO. They go on to share institutional data showing a profound decrease in antithrombin monitoring and replacement per ECMO hour following transition to the DTI bivalirudin as their first line anticoagulant. Use of DTIs in pediatric patients supported by ventricular assist devices (VADs) has increased in use and resulted in the thoughtful creation of a DTI harmonization guide in 2017 as part of the Advanced Cardiac Therapies Improving Outcomes Network (ACTION) in an attempt standardize treatment across institutions.2 A multicenter retrospective review of patients found lower rates of major bleeding and strokes compared to other anticoagulation agents.3 After gaining increased experience with bivalirudin in this population and seeing improved outcomes, our institution made the decision to transition from UFH to bivalirudin as our first line anticoagulant in patients supported by ECMO outside of the neonatal intensive care unit (NICU). Because of concerns related to the use of roller pump circuits and decreased bivalirudin effectiveness in the setting of reservoir stasis, we are yet to make the transition to bivalirudin in the NICU population. We recently reviewed our data for the use of bivalirudin and found no difference in outcomes between UFH and bivalirudin, despite longer ECMO runs in the bivalirudin group.4 In addition, patients anticoagulated with bivalirudin had a longer freedom from first circuit intervention. Although we did not see a difference in outcomes, our other findings are encouraging. It is worth noting that our cohort did not include our NICU population, who could materially benefit from the use of DTIs because antithrombin deficiency is well documented in the neonatal age group.5 There continues to be significant variability in pediatric ECMO anticoagulation, without clear evidence for a first line agent. As Seelhammer et al. clearly convey the advantages of using bivalirudin make it a reasonable replacement for UFH, but multicenter data to support this use is absent. While we found some indicators showing bivalirudin may be superior to UFH in our single center experience, clinical outcomes were not different. It is not unreasonable to believe that a coordinated multi-center approach using bivalirudin as the primary ECMO anticoagulant, similar to the approach taken by ACTION, could show bivalirudin is superior to UFH. This lends the question, is the best yet to come for the use of bivalirudin to anticoagulate patients supported on ECMO.