Bivalirudin in Extracorporeal Life Support: Response and Considerations

To the Editor: We would like to thank Seelhammer et al.1 for their observations regarding our experiences with the use of bivalirudin in extracorporeal life support (ECLS) published in this journal.2 We agree with most of the points made regarding the pharmacokinetic and pharmacodynamic properties of bivalirudin and application to use in ECLS. We do understand and have previously described the issue of proteolytic cleaving and loss of effect in stagnating blood.3 During cardiac surgery and cardiopulmonary bypass performed with concurrent bivalirudin, we add a citrate solution to our cell savers or to any collected blood to prevent clotting.4 In addition, we recognize the potential for nonlinearity and flattening of the activated partial thromboplastin time (aPTT) curve for which we have previously reported.5 We have observed on several occasions an apparent lack of aPTT response despite high doses of bivalirudin (over 0.25 mg/kg/hr), but conversely witnessed a concurrent increase in the international normalized ratio and thrombin time. We additionally were engaged with the authors in the case described by Hellwig et al.6 for argatroban as you noted, which suggested that this may be a situation faced with direct thrombin inhibitors in general. As noted by Seelhammer et al.,1 an alternative approach to assess the level of anticoagulation independent of the aPTT alone may be an important consideration, which we have utilized in our management process. Regarding the initiation of bivalirudin, our institution provides care for a very high acuity population, and at times, our ECLS patients have serious comorbid conditions such as continuous renal replacement, continued open chest postcardiac surgery, and impaired liver or cardiac function. This leads us to a more individualized management process utilizing a small group of clinicians experienced in managing bivalirudin infusions. Our approach to initiating the bivalirudin infusion, while at times removing the heparin therapy and sustaining the desired level of anticoagulation using a revised aPTT target as reagent responses are different, was carefully thought out in advance to anticipate potential effects on measured laboratory parameters and notable bedside observation (e.g., clinical indicators) instead of being reactionary. In cases of high bleeding risks and lower thrombotic concerns, given our experience with notable excessive aPTT response at low doses, we previously described an approach to a stepwise progression of dosing to target aPTT values.7 Although using estimated creatinine clearance is a means to guide therapy, this method can be inaccurate at times, such as in the setting of changing renal function, the presence of renal support therapy, and acute kidney injury with fluctuating function compared with chronic kidney disease.8 Our populations on ECLS are not a homogeneous group, where notable variabilities exist, and this limits the use of an established protocol for initial management. In contrast, a guideline for the approach with the flexibility to individualize therapy may be an alternative consideration. This may include higher-end aPTT goals (e.g., 2–2.5 × baseline value) for low flow, for clotting observed in the oxygenator or tubing, or for clotting occurring on concurrent dialysis circuits. On the other hand, when there are concerns for bleeding and a lower risk of thrombosis, a lower end aPTT target may be considered (e.g., 1.5–2 × baseline value). William E. DagerElizabeth A. WalkerA. Joshua RobertsDepartment of PharmacyUniversity of California Davis Medical CenterSacramento, California[email protected]