Use of a Calibrated, Drug-Specific DTT Assay to Improve Bivalirudin Monitoring in Pediatric Extracorporeal Membrane Oxygenation (ECMO)

Background: Balancing the hemostatic system in anticoagulated ECMO patients to optimally prevent thrombosis and bleeding is challenging. Although the traditional anticoagulant of choice for ECMO has been unfractionated heparin (UFH), use of the direct thrombin inhibitor bivalirudin is growing, due to reported reductions in bleeding and thrombotic complications. Bivalirudin monitoring remains problematic, as the most widely available and commonly used assay, the activated partial thromboplastin time (aPTT), correlates poorly with other assays of bivalirudin effect. An accurate, standardized monitoring test is needed. In a prior study we investigated use of a research-only dilute thrombin time (dTT) assay calibrated to bivalirudin concentration (Hemoclot, Hyphen-Biomed, FR) in a pediatric ventricular assist device (VAD) population. This bivalirudin dTT had excellent correlation with both the clinical dTT (R² = 0.991) and with a research-only, drug-calibrated factor IIa chromogenic assay (BiophenDT, Hyphen-Biomed, FR) (R² = 0.94), but poor correlation with the aPTT (R² =0.1). This led to an institutional practice change to utilize the clinical dTT and the bivalirudin dTT assay for monitoring bivalirudin in VAD patients. Clinicians started to include these tests as well as the aPTT for patients on ECMO; however, validation of the test in the ECMO population has not been reported.

Aim: The aim of this quality improvement project was to validate how the bivalirudin dTT correlated with the clinically ordered dTT (c-dTT) and aPTT in pediatric ECMO patients.

Methods: A retrospective review of pediatric ECMO patients ages 0-18 years treated with bivalirudin was conducted. Bivalirudin dosing, monitoring and titration was at the discretion of the treating physician. Clinically ordered laboratory testing of interest included c-dTT, bivalirudin dTT, and aPTT. The bivalirudin dTT uses a known quantity of human thrombin added to a 1:10 diluted plasma sample to initiate clotting. Clotting time (seconds) is related to bivalirudin concentration using drug-specific controls and calibrators of low and high (1.5 and 4 microgram/mL) bivalirudin (Biophen, Hyphen-Biomed, FR). Clinical data is reported descriptively. The bivalirudin dTT was compared to aPTT and c-dTT results using Pearson correlation and an R² coefficient determined by simple linear regression. Data analysis on clinical outcomes including circuit thrombosis, patient thrombosis, and bleeding events is ongoing.

Results: In total, 124 clinically obtained, concurrent c-dTT, aPTT, and bivalirudin dTT results from 6 pediatric ECMO patients receiving bivalirudin were evaluated. Ages ranged from 0-14 years (mean 3.5 years) with 4 patients started on ECMO in the neonatal period (<1 month of age). Indications for ECMO included congenital heart disease, congenital diaphragmatic hernia, respiratory failure, and cardiogenic shock. Mean bivalirudin infusion rate was 0.6 mg/kg/hr (range 0.15-0.84mg/kg/hr). Average duration on bivalirudin was 8 days (range 4-11) for 5 of the 6 patients, with 1 patient on for > 150 days. The mean aPTT was 78.5 seconds (range 55.9-105.5 seconds). The mean c-dTT was 51 seconds (range 25.4-71.1 seconds). The bivalirudin assay had poor correlation with aPTT (R² = 0.273, Fig 1) and excellent correlation with the c-dTT (R² = 0.899, Fig 2). A c-dTT range of 35-65 seconds corresponded with a bivalirudin dTT level of 0.55-1.8mcg/mL.

Discussion: Here we report results of the clinical use of a bivalirudin dTT monitoring assay in pediatric ECMO patients. Similar to prior results in VADs, the aPTT showed poor correlation with the bivalirudin dTT. The c-dTT had excellent correlation with the bivalirudin dTT and results were so stable for an infusion rate that clinicians ordered less frequent monitoring labs, compared to their practice for UFH. More research and outcomes data is needed to determine the appropriate therapeutic range for optimizing bivalirudin effect and minimizing complications. Validation of a bivalirudin-specific assay that can be used by many institutions will facilitate collaboration on outcomes-based research.

No relevant conflicts of interest to declare.