Since its introduction to bedside clinical practice over 40 years ago, extracorporeal life support (ECLS) has been continually changing and improving as a life-saving technology. Despite this, there remain several areas for potential improvement. Extracorporeal life support disrupts the normal finely maintained balance of coagulation and fibrinolysis by exposing large amounts of blood to nonendothelial surfaces.1 This leads to an inflammatory response with activation of the coagulation cascade and the need for systemic anticoagulation.2 Promising work is being done in the realm of nonthrombogenic biomaterials,3–5 3–5 3–5 but until these are a reality in clinical practice the reliance on systemic anticoagulants remains. Unfractionated heparin (UNFH) is currently the standard anticoagulant in ECLS. Although it has attractive qualities, UNFH is not without its flaws. Alternative anticoagulants have been recently developed with improved safety profiles and reliable monitoring. Within this group of agents are the direct thrombin inhibitors (DTIs) bivalirudin and argatroban. The purpose of this article is to compare these DTIs with the current standard of UNFH anticoagulation during ECLS, to evaluate the current literature surrounding the use of these drugs in ECLS, and finally to propose therapeutic guidelines for their use in ECLS.
Unfractionated heparin is currently the most common method of anticoagulation in ECLS.6 Its effect is mediated by binding antithrombin (AT), an endogenous anticoagulant produced by the liver. The effect of UNFH + AT complex is 1,000× greater than AT alone. The main targets of UNFH + AT are irreversible inhibition of both thrombin and factor Xa, although it also inhibits factors IXa, XIa, and XIIa to a lesser extent.1 Due to a conformational change that hides the binding site, the complex will not inhibit thrombin that is already bound to fibrin,7 thus making it ineffective against pre-existing clot. Reversal of the anticoagulant effect can be accomplished by administration of protamine, and although this makes it an attractive option for cardiopulmonary bypass (CPB), reversal of anticoagulation in ECLS is rarely required.
Unfractionated heparin has a half-life of 60 minutes and is metabolized in the liver as well as the reticuloendothelial system. Pharmacokinetics can be unpredictable due to factors affecting AT activity, variable metabolism rates, and other heparin- neutralizing proteins in the circulation.8 At times it may be necessary to monitor and replace AT with AT concentrate or fresh frozen plasma (FFP). This becomes especially important in infants with who have developmental hemostasis and lower levels of AT.9 Fresh frozen plasma infusion often does not generate adequate concentrations of AT (1 U/ml), therefore AT concentrate is the preferred replacement(1,000–2,000 U/concentrate infusion).10
Monitoring of UNFH effect can be accomplished in a variety of ways. These include measuring activated clotting time (ACT), thromboelastography or thromboelastometry (TEG or ROTEM) in whole blood, or activated partial thromboplastin time (aPTT) in plasma. Heparin infusion is titrated to a given anticoagulant effect, not to a heparin blood level. Activated clotting time is commonly used as it is inexpensive (around $3 per test) and easy to obtain at the bedside. In addition, it measures the coagulation of whole blood including the interaction of heparin with cells and platelets. The goal is to titrate the heparin infusion to maintain a constant low level of anticoagulation. Therefore, the heparin dose and blood level will vary widely during ECLS. Activated clotting time is measured at the bedside by the nurse or extracorporeal membrane oxygenation (ECMO) specialist. If that is not feasible, the alternative is to send citrated blood to the lab to measure accelerated partial thromboplasin time (aPTT). Accelerated partial thromboplasin time is the ACT measured in plasma (and does not account for interactions with platelets and cells) but is still a reasonable measure of anticoagulation effect. In our institution, it costs $60. In our institution, heparin dose is titrated to maintain aPTT at 1.5 to 2 times normal, measured twice a day. Activated clotting time is used during bleeding or clotting complications when heparin dose must be adjusted frequently.
The concentration of heparin in plasma can be measured (indirectly) by the antifactor Xa (antiXa) assay. During ECLS, the heparin dose and concentration will vary widely as heparin is titrated to a given anticoagulant effect (as described above). Therefore, the antiXa assay will be relatively unrelated to heparin effect.9 However, this had led to confusion in the literature of anticoagulation in ECLS because the some investigators have proposed using antiXa levels to titrate heparin. As explained above, heparin (and AT) is titrated to achieve a constant level of anticoagulation, so the antiXa level does not correlate with direct measures of heparin effect (ACT and aPTT).
Thromboelastography or thromboelastometry is a test of the viscoelastic properties of clot formation. It provides information relating to multiple phases of coagulation in the blood that can be especially useful in ECLS. It has been suggested that increased use as a complement to other monitoring tests may help reduce neurologic complications of ECLS2; unfortunately, this technology is not generally available at all institutions and cannot always be done easily when necessary, thus limiting its utility in the management of the critically ill patient. The cost of TEG/ROTEM varies, with ROTEM slightly more expensive, around $15 per test. Protocols of monitoring vary between institutions.
The ECLS organization (ELSO) is the international body that maintains a registry of deidentified data regarding ECLS patients from participating centers. Their proposed guidelines for anticoagulation with heparin recommend the following commonly accepted practice: monitoring ACT (or aPTT) for a goal of 1.5–2 times baseline with intermittent use of other tests if bleeding or clotting complications occur.
One major drawback to the use of UNFH is the development of heparin-induced thrombocytopenia (HIT). This is an immune response involving antibodies that recognize complexes of platelet factor 4 and heparin that can occur in up to 5% of patients receiving heparin therapy. It predisposes the patient to thrombocytopenia and thrombosis. The diagnosis can at times be difficult and relies on a high level of suspicion supplemented by clinical scoring systems, and lab confirmation by serotonin assay.11 False-positive screening assays are common, but the actual condition is rare. The cost of a confirmatory HIT antibody assay is $363 and the serotonin release assay is $332. Treatment entails cessation of heparin with minimal delay and transitioning the patient to an alternative anticoagulant, such as a DTI.
The low cost of heparin is one of the reasons for its continued popularity. At our institution, the cost of UNFH is $0.27 for 1,000 units. Assuming a 70 kg patient that receives a 5,000 unit bolus and is started on an infusion at 18 units/kg/hr, this translates to $9.52 per day. Added expenses can be incurred, however, with measurement of AT at $480 per test. The cost of using heparin with once daily AT test is $490/day. A single 3,000 unit dose of recombinant AT is over $7,000.
Direct Thrombin Inhibitors
Direct thrombin inhibitors are a relatively new class of anticoagulants with several advantages over UNFH. The two commonly used DTIs in clinical practice are bivalirudin and argatroban. They bind to active sites on thrombin (circulating and clot bound)7 increasing their efficacy compared with heparin. In addition, there is no reliance on AT for its effect and therefore allows for more consistent use across all age groups and populations.12 Their selective binding to thrombin and not to other circulating plasma proteins makes pharmacokinetics more predictable. In a recent survey of the ELSO database, DTIs were found to be used by up to 50% of respondents when indicated.6 Indications often included HIT, heparin resistance, and development of thrombosis while on heparin therapy.
As with all new medications, there are always complications and drawbacks to their use. These would include a higher cost than UNFH, no specific antidote or reversal agent, and potential destabilization of existing clot within the patient. The lack of a reversal agent may limit their use in CPB, however would likely not affect their use in ECLS. Their effect on clot-bound thrombin could become an issue in some patients but so is the use of UNFH in certain critically ill patients who require ECLS. It is the judicious monitoring and administration of the medication that will make its use safe.
As stated above, the two most common DTI’s used in clinical practice at this time and more specifically in ECLS are bivalirudin and argatroban.
Bivalirudin is a DTI with a half-life of 25 minutes. It undergoes proteolytic degeneration, leaving metabolism completely independent of the liver and kidney. There are reports, however, that found dose reduction helpful in patients with kidney failure.12 In addition, in patients with cardiac standstill or in areas of stasis, anticoagulation eventually resolves and can result in thrombus formation due to first order metabolism. Adverse effects aside from bleeding include hypotension or hypertension, bradycardia, nausea, headache, and urinary retention. Monitoring of bivalirudin is most often accomplished by following aPTT with a goal 1.5–2× baseline or ACT > 2.5× baseline. Monitoring of DTIs with aPTT and ACT is the most common practice; however, ecarin clotting time (ECT) has been shown to be a superior measure in small studies.13,14 Limited availability has kept ECT from being a more widely used measure.
At our institution, the cost of bivalirudin (Angiomax) is $3.37/mg. Assuming a 70 kg adult receiving a 0.75 mg/kg bolus and a continuous infusion of 0.1 mg/kg/hr, the cost per day is $734.09. In the pediatric literature, the largest case series with 12 patients found the cost was to be $13.68/day for bivalirudin vs. 0.46/day for heparin without antiXa measurement.15 Although not validated, this report also describes possible reversal with rFVII.
There are multiple retrospective reports describing the safety and efficacy of bivalirudin. In one report, 10 children were treated for thromboembolism with bivalirudin. Follow-up ultrasound at 72 hours showed thrombus regression in 10 of 10 patients.16 Pieri et al.17 retrospectively reviewed 20 adult patients on ECLS, 10 patients treated with heparin, and 10 patients with bivalirudin. They found significantly higher variations in aPTT in the heparin group and no differences between bleeding or thrombosis. Ranucci et al. reviewed 21 patients (10 pediatric) on ECLS after cardiac operations, eight of which were on heparin and 13 on bivalirudin. They found statistically more bleeding (as measured by chest drain output), FFP, and platelet transfusions in the heparin group. When factoring in the complications, there was a significantly lower total cost of care in the pediatric group treated with bivalirudin.12 The same group wrote a follow-up cautioning against the use of bivalirudin in circuits with areas of blood stagnation. One patient was found to develop a spontaneous intracardiac thrombus,18 although this also happens with heparin when intracardiac blood is stagnant. There are published doses among pediatric patients, although no validated recommendations as of yet in the literature (Table 1).
Argatroban is an L-arginine derivative with a half-life of 15 minutes. It undergoes hepatic metabolism. Adverse reactions aside from bleeding can include nausea and vomiting, fever, headache, cardiac abnormalities (angina, myocardial infarction, arrhythmia), and infection. Constipation and hypokalemia have also been reported in the pediatric population. Monitoring of argatroban is accomplished by following aPTT for 1.5–2.5× baseline, ACT with a goal of 150–200 seconds, or anti-IIa if available.19 Our institution has very limited experience with argatroban use in ECLS. In general, our pharmacy recommends starting at a reduced dose used in critically ill patients.
The cost of argatroban at our institution is $3.32/mg. Assuming a 70 kg patient on a continuous infusion of 0.5 mcg/kg/min, the cost per day is $167.33.
Argatroban is the DTI most often cited in ECLS applications; however, there are fewer retrospective reports specific to ECLS when compared with bivalirudin. Cornell published a case series of five adult patients on ECLS without any complications noted.20 Beiderlinden retrospectively reviewed nine adult patients on ECLS for ARDS being treated with argatroban after a diagnosis of HIT.21 One critically ill patient had major bleeding and required a dose reduction, thought likely due to impaired hepatic metabolism. The ELSO anticoagulation guidelines set forth recommendations to start at 0.5–1 mcg/kg/min either with or without a bolus and adjust to maintain aPTT 1.5–2.5× baseline.10
The ideal ECLS circuit would not require anticoagulation at all, but for the time being is a necessity. Although UNFH is the current standard of care for anticoagulation during ECLS, it has many flaws. The well-described variability with heparin dosing and metabolism leaves room for improvement. Direct thrombin inhibitors show great promise in ECLS with rapid on/off and ease of use as well as a superior safety profile. Multiple reviews have shown safety and efficacy of DTIs in ECLS. In addition, the added benefit of directly inhibiting uncontrolled thrombin generation and the multifactorial effects it has on already critically ill patients may have added benefits aside from anticoagulant effects. Although increased cost is an argument against DTIs, if taken in context of complications and additional tests and medications that need to be obtained, the overall cost is likely may be comparable with UNFH. Increased use of these medications is needed to establish strong guidelines for their dosing in ECLS.
In cardiac or vascular surgery heparin is ideal because anticoagulation can be complete with a high dose, and then quickly reversed with protamine. Direct thrombin inhibitors cannot be easily reversed and are not used in surgical procedures because of ongoing bleeding. Heparin was used during the development of ECLS as an extension of CPB. Reversal of heparin with protamine is not needed during ECLS, but the use of heparin persists. The problems associated with heparin and related monitoring is significant and may be alleviated by using DTIs instead. This review suggests that DTIs can be used for ECLS, with more reliable and predictable anticoagulation at potentially equivalent cost compared with heparin when monitoring and complications are considered.
Should ECLS centers move to DTI anticoagulation, and what would be involved? There is ample published data to demonstrate the safety and effectiveness as an anticoagulant. In fact, most centers are using argatroban or bivalrudin for cases with positive HIT assays. So it is possible, a center could make a decision to use a DTI for routine ECLS anticoagulation, as we might chose to a different sedative, antibiotic, or inotrope. Anticoagulant effect is monitored as discussed above.
Comparing anticoagulation with DTI to prior experience with heparin would be possible using the ELSO registry. The types of anticoagulation and bleeding/clotting complications are included in the ELSO registry, so a comparison of heparin to DTI can be done on the deidentified single or multicenter ELSO data. A prospective study could strengthen the argument for DTIs in ECLS.
Direct thrombin inhibitors are used for anticoagulation during ECLS in many centers. Direct thrombin inhibitors have been reported to have less clotting and bleeding complications than heparin. Based on the existing experience, it appears that routine use of DTIs for ECLS has the potential for anticoagulation that is more reproducible, easier to monitor, and not more expensive compared with heparin.
1. Annich G, Lynch W, MacLaren G, Wilson JM, Bartlet RHECMO; extracorporeal cardiopulmonary support in critical care. Ann Arbor, MI: Extracorporeal Life Support Organization. 2012
2. Oliver WC. Anticoagulation
and coagulation management for ECMO. Semin Cardiothorac Vasc Anesth. 2009;13:154–175
3. Brisbois EJ, Handa H, Major TC, Bartlett RH, Meyerhoff ME. Long-term nitric oxide release and elevated temperature stability with S-nitroso-N-acetylpenicillamine (SNAP)-doped Elast-eon E2As polymer. Biomaterials. 2013;34:6957–6966
4. Major TC, Brant DO, Burney CP, et al. The hemocompatibility of a nitric oxide generating polymer that catalyzes S-nitrosothiol decomposition in an extracorporeal circulation model. Biomaterials. 2011;32:5957–5969
5. Amoako KA, Archangeli C, Handa H, et al. Thromboresistance characterization of extruded nitric oxide-releasing silicone catheters. ASAIO J. 2012;58:238–246
6. Bembea MM, Annich G, Rycus P, Oldenburg G, Berkowitz I, Pronovost P. Variability in anticoagulation
management of patients on extracorporeal membrane oxygenation: An international survey. Pediatr Crit Care Med. 2013;14:e77–e84
7. Weitz JI, Hudoba M, Massel D, Maraganore J, Hirsh J. Clot-bound thrombin is protected from inhibition by heparin
-antithrombin III but is susceptible to inactivation by antithrombin III-independent inhibitors. J Clin Invest. 1990;86:385–391
8. Young G, Yonekawa KE, Nakagawa P, Nugent DJ. Argatroban as an alternative to heparin
in extracorporeal membrane oxygenation circuits. Perfusion. 2004;19:283–288
9. Bembea MM, Schwartz JM, Shah N, et al. Anticoagulation
monitoring during pediatric extracorporeal membrane oxygenation. ASAIO J. 2013;59:63–68
10. Laurance Lequier GA, Omar A-I, Melania B, et al. ELSO Anticoagulation
Guideline. 2014 Ann Arbor, MI ELSO:3–9
11. Cuker A. Clinical and laboratory diagnosis of heparin
-induced thrombocytopenia: An integrated approach. Semin Thromb Hemost. 2014;40:106–114
12. Ranucci M, Ballotta A, Kandil H, et al.Surgical and Clinical Outcome Research Group. Bivalirudin-based versus conventional heparin anticoagulation
for postcardiotomy extracorporeal membrane oxygenation. Crit Care. 2011;15:R275
13. Schaden E, Kozek-Langenecker SA. Direct thrombin inhibitors: pharmacology and application in intensive care medicine. Intensive Care Med. 2010;36:1127–1137
14. Lind SE, Boyle ME, Fisher S, Ishimoto J, Trujillo TC, Kiser TH. Comparison of the aPTT with alternative tests for monitoring direct thrombin inhibitors in patient samples. Am J Clin Pathol. 2014;141:665–674
15. Nagle EL, Dager WE, Duby JJ, et al. Bivalirudin in pediatric patients maintained on extracorporeal life support. Pediatr Crit Care Med. 2013;14:e182–e188
16. Rayapudi S, Torres A Jr, Deshpande GG, et al. Bivalirudin for anticoagulation
in children. Pediatr Blood Cancer. 2008;51:798–801
17. Pieri M, Agracheva N, Bonaveglio E, et al. Bivalirudin versus heparin
as an anticoagulant during extracorporeal membrane oxygenation: A case-control study. J Cardiothorac Vasc Anesth. 2013;27:30–34
18. Ranucci M. Bivalirudin and post-cardiotomy ECMO: A word of caution. Crit Care. 2012;16:427
19. Chan VH, Monagle P, Massicotte P, Chan AK. Novel paediatric anticoagulants: A review of the current literature. Blood Coagul Fibrinolysis. 2010;21:144–151
20. Cornell T, Wyrick P, Fleming G, et al. A case series describing the use of argatroban in patients on extracorporeal circulation. ASAIO J. 2007;53:460–463
21. Beiderlinden M, Treschan T, Görlinger K, Peters J. Argatroban in extracorporeal membrane oxygenation. Artif Organs. 2007;31:461–465