Bivalirudin Monitoring By Dilute Thrombin Time Is Cost-Efficient in Pediatric ECMO Patients

Background: Use of the parenteral anticoagulant bivalirudin, a direct thrombin inhibitor, is increasing for pediatric patients requiring mechanical circulatory support (MCS) because of its reported favorable safety (bleeding, blood transfusions) and efficacy (thrombosis) outcomes, compared to unfractionated heparin (UFH). Recent publications have documented favorable or equivalent clinical costs for patients on MCS who are anticoagulated with bivalirudin vs UFH, however these publications have included the cost of the activated partial thromboplastin time (aPTT) for bivalirudin monitoring. In previous publications, our group has demonstrated that the aPTT is an unreliable marker of bivalirudin’s anticoagulant effect in both extracorporeal membrane oxygenation (ECMO) and ventricular assist device (VAD) patients; the dilute thrombin time (dTT) provides superior reliability. Furthermore, a drug-calibrated dTT could offer the opportunity for multicenter collaboration to identify a true therapeutic range for bivalirudin anticoagulation using relevant clinical outcomes. Concern remains among clinicians that the dTT is largely unavailable for use in monitoring bivalirudin due to access and/or cost issues at their institutions. Here, we report a cost analysis of monitoring ECMO patients anticoagulated with bivalirudin versus UFH.

Method: As part of a larger, IRB-approved, retrospective clinical study, we collected clinical and laboratory data for children admitted to Intensive Care Units at Cincinnati Children’s Hospital and Medical Center on ECMO and anticoagulated with UFH or bivalirudin between January 16, 2018 and August 30, 2023. We performed cost analysis by identifying the total number of laboratory tests (CBC, aPTT, fibrinogen, AT3, UFH level, dTT, and free Hgb) ordered for patients during the first 5 days of ECMO and used the 2024 institutional list prices for each test.

Results: A total of 47 patients anticoagulated with UFH and 30 patients anticoagulated with bivalirudin were included for analysis. The average number of tests drawn in the bivalirudin group were decreased, specifically: complete blood count (UFH versus bivalirudin, 19.6 vs 17.7 tests, p=0.012), fibrinogen (18.0 vs 14.8, p=0.003), unfractionated heparin level (19.0 vs 7.1, p<0.001), and antithrombin (15.6 vs 6.3, p<0.001). The average number of monitoring labs per day of ECMO (UFH level for the UFH group and dTT for the bivalirudin group) was similar within the first five days of ECMO support – however, with UFH cost ($211) vs dTT ($126) and fewer antithrombin levels ($418) needed in the bivalirudin group, this represents further cost savings for bivalirudin-monitored patients. Finally, average monitoring cost per day was decreased in the bivalirudin group ($2164 vs $3226 in the UFH group).

Discussion: The dTT, previously demonstrated as superior to the aPTT for monitoring bivalirudin therapy, can be performed in clinical laboratories on an automated or semi-automated platform. Clinical guidelines at many institutions (including Extracorporeal Life Support Organization [ELSO] members) continue to use aPTT for bivalirudin monitoring with concerns for cost and lack of expertise/availability for dTT in their clinical laboratories; at our ELSO member institution, multiple daily aPTTs are ordered, even as dTT is clinically used for managing bivalirudin. Monitoring cost and blood sampling volume savings will be enhanced by reducing aPTT orders to only clinically necessary. Even so, these data demonstrate that bivalirudin monitoring by dTT in MCS is cost-efficient for the main laboratory parameters tested. Previous studies have reported decreased need for blood product transfusions and circuit changes with bivalirudin, with favorable or equivalent total costs in MCS patients. This work demonstrates that dTT monitoring of bivalirudin in pediatric patients during the first 5 days of ECMO is cost-efficient and potentially reduces total laboratory test/blood volume sampling.