Extracorporeal membrane oxygenation (ECMO) is a life-saving intervention that provides gas exchange and circulatory hemodynamic support to patients with severe respiratory and/or cardiac failure. However, continuous blood exposure to artificial surfaces of ECMO circuits precipitates an inflammatory response, coagulation factor activation, and an acquired von Willebrand deficiency 1. Thus, the coagulation derangements in critically ill patients are in a state of perpetual flex, making hemostatic management during ECMO support resemble an ongoing quest for the challenging balance between preventing thrombotic complications while simultaneously minimizing bleeding events, which is progressively becoming a more frequent complication 2. Accordingly, hemostatic complications remain the leading contributors to morbidity and mortality in patients supported on ECMO, making consistent and reliable anticoagulation critically important 3. Central to successful ECMO support is maintenance of satisfactory membrane lung function. Thus, the Extracorporeal Life Support Organization recommends routine anticoagulation during ECMO support 4. Although they make no specific recommendation with regards to choice of agent, unfractionated heparin is the predominantly used anticoagulant for ECMO support worldwide, although this is not recorded in the registry limiting the ability to perform global epidemiological studies 4. However, despite heparin anticoagulation, oxygenator failure requiring circuit exchanges remains unacceptably high with rates typically in the 10% to 30% range and some studies reporting occurrence in upward of 80% to 100% of all ECMO runs 5. Notable limitations of heparin may serve as explanations for its suboptimal performance in preserving the membrane lung during ECMO support, including its requirement for antithrombin as a cofactor, variable responsiveness due to heparin resistance, susceptibility to platelet factor 4 interactions and immune-mediated thrombocytopenia, extensive binding to circulating plasma proteins, and batch-to-batch potency variation inherent to biologically derived products 6. Although membrane lungs tend to be somewhat resilient, excessive deposition of thrombotic material leads to increased blood resistance and impairs acceptable gas exchange 7. Although ECMO circuits can be replaced in such contexts, disposables are high-priced items with a history of stocking issues, and these procedures may be accompanied by complications including cardiac arrest. Thus, a more effective preventative strategy to prolong the life of membrane lungs should be prioritized. To this end, bivalirudin is becoming an increasingly popular choice for maintenance anticoagulation during ECMO support, even in the absence of heparin-induced thrombocytopenia. Meta-analyses of empirical evidence suggest bivalirudin to be superior to heparin in patients supported by ECMO, particularly at reducing circuit-related thrombotic events 6. Potential reasons for these differences are likely multifactorial, but may include bivalirudin having no plasma protein interaction, no reliance on a cofactor, and that it produces generally more reliable dosing than heparin that results in a greater time spent within therapeutic anticoagulation targets during ECMO 8. Key pharmacological differences in an anticoagulant's interactions with thrombin may further influence its ability to achieve optimal membrane lung preservation during ECMO support. Specifically, bivalirudin's binding site on thrombin remains exposed when thrombin is bound to clot, whereas the heparin-antithrombin binding site becomes inaccessible once thrombin is clot-bound 9. Thus, although the heparin-antithrombin complex only engages with freely circulating thrombin, bivalirudin can engage with and inhibit not only circulating thrombin but also thrombin that is already bound to fibrin on established clot. Additionally, bivalirudin effectively suppresses thrombin-dependent platelet activation via inhibition of protease-activated receptor 1 and 4 cleavage thereby inhibiting collagen-induced platelet procoagulant activity and modulating the crucial first step in platelet activation. Although seemingly esoteric, these pharmacologic distinctions represent attractive features for anticoagulating ECMO circuits, particularly given the high propensity for thrombotic deposition on membrane fibers 1. Indeed, studies have reported utilizing catheter-directed bivalirudin infusion locally to areas of thrombus formation to achieve a thrombolytic-type effect 10. In 2017, the Mayo Clinic ECMO program introduced bivalirudin as a preferred anticoagulant option for adult ECMO support. Contraindications to its use included very low flow states, minimal or absent left ventricular pulsatility, or conditions impacting the reliability of aPTT monitoring such as lupus anticoagulant. Ultimately, the decision to use anticoagulation and the choice of agent were at the discretion of the treating clinicians. All adult patients have been supported by the CardioHelp system (Getinge, Gothenburg, Sweden) with the HLS Set Advanced 5.0 or 7.0 and a QUADROX family oxygenator (Maquet, Getinge Group, Gothenburg, Sweden). All patients receive 50 to 100 units/kg bolus of intravenous heparin for ECMO cannulation. Systemic anticoagulation is initiated thereafter once hemostasis is deemed satisfactory. Bivalirudin is initiated at a rate of 0.02 to 0.15 mg/kg/h depending on the underlying kidney function or presence of renal replacement therapy. The aPTT is monitored every 6 h and adjusted to the typical range of 60 to 80 s. We have previously published the fully detailed protocol 11. In our institutional experience, we infuse bivalirudin directly into the ECMO circuit via a dedicated pigtail bridge at the top of the oxygenator (Figure 1). The primary goal of this was to maximize the regional concentration of bivalirudin thereby increasing therapeutic efficacy at the local site of contact phase activation that drives thrombin generation in the context of extracorporeal support. The infusion is initiated toward the pre-oxygenator side to ensure any potential air is captured before reaching the patient. Every 4 h, during the ECMO specialist's circuit assessment, the stopcock is turned to alternate flow between the pre- and post-oxygenator sides to maintain patency of both pigtails. All infusions have in-line air filters and are carefully primed before connection. Stopping or restarting follows standard intravenous precautions (i.e., clamp, re-prime, and confirm no air), whereas the specialist verifies pressures and circuit integrity before resuming. Since introducing bivalirudin to our ECMO program, we have supported over 600 ECMO runs with this bivalirudin administration mechanism (Figure 2). Although bivalirudin is our preferred agent, the number of supported runs relatively decreased in recent years as some clinicians may prefer heparin in certain situations, as well as increasing use of an “anticoagulation-free” strategy in select cases. The bivalirudin administration mechanism has remained consistent and safe over this time, and we have had zero safety events related to this practice. Actually, the Extracorporeal Life Support General Guidelines (current v1.4) state that it is acceptable to access the circuit for monitoring, sampling, and infusions, provided protocols are present to do so safely. Indeed, others have reported safe pre-oxygenator site heparin infusion during flow reduction trials 7. Importantly, we did not have access to a control group of patients who received bivalirudin through a systemic catheter (not ECMO). Therefore, efficacy compared with standard delivery mechanisms should not be inferred from our report, but rather that the technique can be done so safely with proper protocols in place. In summary, we posit that if bivalirudin is used for routine maintenance ECMO anticoagulation, consideration should be made for directly infusing into a pre-membrane access site. Doing so with bivalirudin may further enhance its pharmacodynamic effectiveness to preserve the membrane lung. However, the location of membrane access sites may influence the distribution and saturation of fluid infused. Although the pre-membrane access site on the QUADROX oxygenator we used is at the top of the device above the inlet blood flow, fluid infused may not reach the entire network of fibers. It is possible that other oxygenator devices that have different access site designs with the pre-membrane access site directly in the inlet blood flow path may be more optimal for bivalirudin infusion. In silico and ex vivo studies evaluating fluid dynamics and membrane saturation of bivalirudin infused through pre-membrane access sites are needed to further elucidate the optimal strategy for delivering anticoagulation during ECMO for maximum membrane lung preservation. P.M.W. and T.G.S. conceived the editorial. P.M.W. created the visuals, provided supervision, and drafted the first version of the manuscript. All authors reviewed and critically revised the manuscript for important intellectual content. All authors reviewed and approved the final version of the manuscript. The authors have nothing to report. The authors have nothing to report. P.M.W. receives consulting fees from Wolters Kluwer UpToDate. All other authors report no conflicts of interest. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.
Wieruszewski et al. (Thu,) studied this question.