Anticoagulation Strategies during ECMO Application

Abstract

During ECMO support, optimal anticoagulant drugs, dosing charts, ideal anticoagulation levels, and monitoring parameters have not yet been definitively established, despite the increasing use of ECMO applications worldwide. Heparin remains a widely used anticoagulant, despite its age and known limitations. While interest in direct thrombin inhibitors is growing, dosage and safety information are still limited. Presently, there is a trend toward combining traditional or newer anticoagulant drug usage with modern technological advancements to manage coagulation disorders more effectively and safely. Achieving optimal anticoagulation during ECMO involves leveraging a multidisciplinary approach that integrates pharmacokinetics and personalized dosing algorithms. The management of anticoagulation should be individualized for each patient, taking into account their specific characteristics, clinical condition, and laboratory results. Treatment plans are tailored based on an individual’s genetic predisposition to clotting and their response to anticoagulants, with the aim of minimizing adverse effects and optimizing therapeutic outcomes. Continuous and real-time assessment of the coagulation status enables timely and appropriate anticoagulation therapy. The integration of cutting-edge technologies such as artificial intelligence and machine learning may enhance the overall safety profile of anticoagulation treatment during ECMO. Advancements in anticoagulant therapy in ECMO continue to progress. This approach, utilizing genetic information, real-time monitoring, and advanced technologies, aims to provide an individually optimized treatment strategy for the management of coagulation disorders.

1. Introduction

ECMO applications are increasingly becoming widespread worldwide. The success and sustainability of these applications depend on the appropriate selection of anticoagulant drugs and the effective management of anticoagulation during ECMO support. However, the recommended anticoagulant drugs, dosage tables, monitoring parameters, and ideal anticoagulation levels during ECMO support have not yet been definitively established. Despite the widespread use of heparin as an anticoagulant, the search for the ideal anticoagulant drug continues due to its deficiencies and the risk of heparin-induced thrombocytopenia.

Direct thrombin inhibitors offer advantages over heparin due to their ability to provide more stable anticoagulation levels, inhibit both bound and circulating thrombin and lack the risk of heparin-induced thrombocytopenia. However, they do not have antidotes. Consequently, in some ECMO centers, direct thrombin inhibitors are now being preferred over heparin during ECMO support due to their positive effects.

2. Extracorporeal membrane oxygenation and anticoagulation

2.1 Historical overview

Successful extracorporeal membrane oxygenation (ECMO) support was first applied in an adult patient with advanced respiratory failure due to severe lung contusion in 1972 [1]. Since the 1980s, extracorporeal life support has emerged as a significant therapeutic option in critically ill patients, and ECMO utilization has progressively increased since 2010. Advancements in oxygenator technology, particularly the use of nano-porous hollow fiber membranes (PMP), have led to reduced trauma and thromboembolism in blood components, consequently decreasing device-related bleeding complications. In 2022, ELSO reported its 200,000th ECMO patient [2].

Studies on anticoagulant drugs began as early as 1916, with the anticoagulant developed in 1918 being named heparin by Howell and Holt [3]. In 1966, Wardrop introduced activated clotting time (ACT) to monitor the anticoagulant effect of heparin. Subsequently, ACT monitoring during cardiac surgery became the primary method for assessing anticoagulation. From the 1970s onwards, the clinical use of heparin as an anticoagulant steadily increased, marking a significant milestone in cardiac surgery.

The crystal structure of human thrombin was elucidated in 1989, leading to research on direct thrombin inhibitors (DTIs), with argatroban being the first developed. Following this, numerous DTIs have been developed.

2.2 ECMO and anticoagulation

The initiation of ECMO leads to the contact of blood with foreign surfaces, triggering an immune system response in the form of inflammation and coagulation cascades. The initiation of ECMO is associated with an inflammatory reaction similar to that observed in systemic inflammatory response syndrome (SIRS). The coagulation cascade is activated, leading to systemic inflammatory response through the activation of the complement systems, endothelial cells, leukocytes, and platelets. This response may be analogous to the response observed during cardiopulmonary bypass. Activation of the contact system triggers both intrinsic and extrinsic coagulation pathways, leading to thrombus formation. Activated factor X (FXa) converts prothrombin to thrombin, which subsequently converts fibrinogen to fibrin, resulting in thrombus formation [4].

Additionally, during ECMO, there is time-dependent platelet activation, accompanied by a decrease in platelet aggregation [5].

During cardiopulmonary bypass, the inflammatory response has been extensively studied, but the inflammatory response during ECMO has been less investigated. The details of the coagulation and inflammation activated during ECMO have not yet been fully elucidated. Critical illness, often associated with sepsis, triggers an inflammatory response, which becomes more complex when ECMO is applied to these patients [4].

Anticoagulation during ECMO has been described as a dynamic process frequently interrupted in cases of bleeding [6]. During ECMO, the bleeding rate is reported to be 16.5%, with at least one bleeding event occurring in 52.5% of patients. Elevated activated partial thromboplastin time (aPTT) is a potential risk factor for the initial bleeding episode.

To prevent bleeding during surgery, ECMO can be used without anticoagulation by individually assessing the patient, utilizing heparin coating of the tube system. The success of ECMO applications critically depends on applying anticoagulation to keep the patient’s arterial and venous systems, as well as the ECMO circuit, free from thrombi without inducing bleeding. Both critical illness and inflammatory responses associated with ECMO complicate achieving the desired level of anticoagulation, making it challenging and complex.

If intracranial hemorrhage is present, anticoagulation may not be used during VV ECMO [7]. In patients with bleeding diathesis or intracranial hemorrhage, it is important to be prepared for ECMO oxygenator exchange if necessary.

Concerns regarding the prevention of thrombosis in ECMO circuits may lead to undesired situations where active systemic anticoagulation is not desired, especially in patients with a high risk of bleeding. Due to the increased mortality associated with bleeding during ECMO, there has been a tendency toward the adoption of a protective anticoagulation strategy with heparin. In a case series involving 40 patients undergoing VV-ECMO treatment, a comparison was made between a protective anticoagulation strategy with a target ACT of 140–180 seconds and a full therapeutic anticoagulation strategy with a target ACT of 180–200 seconds. Multivariable analysis did not reveal statistically significant differences in terms of mortality, bleeding risk, or thrombotic complication risk. It is noteworthy that 35% of ECMO circuit days across all patients were heparin-free. Survival rates, bleeding incidents, thrombotic complications, and transfusion requirements did not differ between the protective and full therapeutic heparin strategies in the management of VV ECMO. However, the limitations of this study include its inclusion of a limited number of patients and the potential confounding effects of including patients who utilized both single-lumen and double-lumen cannulas [8].

The application of prophylactic subcutaneous anticoagulation without systemic anticoagulation has been reported in VV-ECMO procedures. In a single-center observational study, thrombosis prophylaxis was conducted using 40 mg of subcutaneous enoxaparin. A centrifugal pump observed 560 ECMO days, and in four patients, the pump stopped due to thrombotic occlusion after five days. Significant clinical bleeding was observed in only 18% of patients. It was reported that anticoagulation at prophylactic dosages was feasible and not associated with thrombosis when applied in patients without thrombosis [9].

A systematic review examining ECMO applications without systemic anticoagulation analyzed two prospective studies and others consisting of case reports. In this review, it was found that in patients without systemic anticoagulation, the incidence of circuit and patient thrombosis was similar to those who received continuous systemic anticoagulation. The median duration of ECMO without systemic anticoagulation was 4.75 days. Additionally, this review reported a case of ECMO followed for 130 days without anticoagulation [10].

In pediatric patients at high risk of bleeding, ECMO was utilized without systemic anticoagulation, resulting in a lower frequency of patient or circuit thrombosis. A total of 35 patients were followed for 964 hours of ECMO support [11].

Optimal anticoagulant drugs, dosage tables, ideal anticoagulation levels, and monitoring parameters during ECMO support have not yet been definitively established, despite the increasing use of ECMO worldwide. There is still limited international consensus on this matter, and anticoagulation management relies on updates from the Extracorporeal Life Support Organization (ELSO) [8]. Generally, centers have their own practices or protocols regarding anticoagulation management during ECMO.

The anticoagulant drugs currently in use can be categorized into three main groups:

• The heparin group,

• Direct thrombin inhibitors, and

• The Nafamostat group.

Heparin stands as the initial and most frequently utilized anticoagulant medication. Nonetheless, it presents several deficiencies as an anticoagulant.

Direct thrombin inhibitors have garnered increased significance in contemporary practice. Among these anticoagulants, bivalirudin, specifically during ECMO utilization, exhibits advantages over heparin and has begun to supplant it. However, information regarding the dosage and safety profiles of this anticoagulant group remains limited.

In contemporary practice, there is a tendency toward a new approach that combines the use of traditional or newer anticoagulant drugs with modern technological advancements to enhance the effectiveness and safety of anticoagulation management. There has been a growing focus on individualized anticoagulant management through the application of machine learning and monitoring individual responses to anticoagulants.

The Extracorporeal Life Support Organization (2021) guidelines emphasize the need for individualization of anticoagulation [12]. Existing anticoagulation tests are unable to assess all stages of coagulation and their effects on fibrin formation. The clinical goal is to achieve complication-free ECMO management.

2.3 The commonly used anticoagulant drugs during ECMO

2.3.1 Heparin

Heparin remains one of the oldest and most frequently used anticoagulants despite its shortcomings. It belongs to the class of glycosaminoglycans (GAGs). Unfractionated heparin (UFH) is a heterogeneous mixture of heparin molecules ranging from 5 to 25 kDa. Its antidote is protamine. Most heparins used are purified from porcine and bovine intestinal mucosa, and to a lesser extent, from bovine lungs. Heparin tightly binds to a specific region of antithrombin, catalyzing the inhibition of thrombin indirectly [13].

Binding of heparin to antithrombin makes antithrombin twofold more reactive against thrombin and 100-fold more reactive against FXa. Heparin needs to bind to antithrombin for thrombin inhibition. The heparin-antithrombin complex cannot bind to fibrin-bound thrombin and is protected from inhibition by heparin [14]. However, only one-third of heparin molecules can bind to antithrombin (AT). Heparin also possesses procoagulant/antifibrinolytic properties. These include stimulation of protein C inhibition by protein C inhibitors, altered inactivation of factor Va, stimulation of clotting initiated by FXa, and weakening of fibrinolytic pathways [5]. Heparin has long been used for anticoagulation during cardiac surgery. Unfractionated heparin is the first and most commonly used anticoagulant in 77% of patients during ECMO, with the highest usage rate at 73% [6]. Nevertheless, significant variability exists among patients in heparin dosage and laboratory parameters and values monitored for anticoagulation.

While heparin has been used for a long time as an anticoagulant, it does have its limitations. It does not have any effect on thrombin formation or the inhibition of thrombin-fibrin complexes. It becomes ineffective after thrombin formation. Heparin binds to plasma proteins, proteins released from platelets, and endothelial cells, leading to a variable anticoagulant response. It cannot render factor Xa in the prothrombinase complex or thrombins bound to fibrin or subendothelial surfaces inactive [15].

Heparin resistance refers to the diminished response to heparin anticoagulation and is a drawback of heparin use. Antithrombin deficiency is the primary mechanism of heparin resistance. Heparin resistance varies from 4 to 26% during cardiopulmonary bypass. When heparin resistance is detected, antithrombin concentrates, fresh frozen plasma, additional heparin, or nonheparin anticoagulants such as direct thrombin inhibitors like bivalirudin may be selected.

Heparin-induced thrombocytopenia (HIT) is another undesirable effect associated with heparin. The incidence of HIT in patients receiving UFH treatment for more than five days is approximately 3% [16]. The formation of IgG antibodies against the platelet factor 4/heparin complex leads to immune-mediated destruction and simultaneous activation of platelets, resulting in hypercoagulability and venous and arterial thrombosis (such as deep vein thrombosis, pulmonary embolism, and skin necrosis). Immune-mediated thrombocytopenia (aHIT) is a subgroup of HIT and can be life-threatening. Anti-platelet factor 4 (anti-PF4) antibodies activate platelets and lead to thrombosis. Clinical monitoring, including platelet counts and antithrombin III levels, as well as identification of heparin resistance and HIT using thromboelastography, can aid in diagnosis [17]. Anti-PF4 antibodies can develop independently of heparin [17]. Immediate discontinuation of heparin and initiation of nonheparin anticoagulation is the standard treatment for HIT. In cases of treatment-resistant HIT, alternative therapies such as plasma exchange and intravenous immunoglobulin may be required [18]. To avoid HIT, heparin should be replaced with fondaparinux or a direct thrombin inhibitor. Heparin-induced skin necrosis can occur, leading to high morbidity and rare mortality.

The binding of unfractionated heparin (UFH) to circulating positively charged plasma proteins (such as von Willebrand factor and C-reactive protein), endothelial cells, and macrophages can alter the pharmacokinetics of UFH and the patient’s dosage requirements. This diminishes the anticoagulant effect of heparin and is a significant drawback of heparin use.

In neonatal and pediatric patients, pharmacokinetics differ significantly due to lower antithrombin concentrations and larger volume of distribution. The half-life of heparin is 60–90 minutes in adults, whereas it ranges from 35 to 70 minutes in pediatric patients.

The ACT is still widely used for monitoring the anticoagulant effect of heparin, especially during cardiac surgery, as it can be easily assessed at the bedside in a short time. However, ACT is influenced by various factors such as hypothermia and hemodilution. Therefore, it remains uncertain whether a decrease in heparin response measured by ACT indicates inadequate anticoagulation. During cardiopulmonary bypass, ACT is primarily used, and despite target values of 400–480 seconds, ACT levels can vary between 240 and 1000 seconds. Although activated aPTT provides slower results compared to ACT, it is more suitable for assessing heparin response, especially during ECMO, and is often monitored alongside ACT. During ECMO, the target value for aPTT should be 1.5–2 times the patient’s baseline value.

The optimal monitoring strategy for heparin and a well-established dose that offers good safety and efficacy profiles have not been definitively determined. Viscoelastic clotting tests have not been compared in terms of clotting time or heparin concentration.

None of the tests used for anticoagulation evaluation alone can fully assess the adequacy of anticoagulation. Therefore, there is a trend toward evaluating anticoagulation levels with bedside viscoelastic tests within the capabilities of healthcare centers.

In addition to the risk of HIT and the shortcomings of heparin as an anticoagulant, the contamination crisis during heparin use in 2008 has led to investigations into nonanimal-derived heparin-like anticoagulant sources [19].

2.3.2 Direct thrombin inhibitors (DTIs)

Thrombin plays a crucial role in thrombus formation. After its formation, thrombin converts soluble fibrinogen into insoluble fibrin and stimulates platelet activation. DTIs bind directly to thrombin without requiring a cofactor like antithrombin. They can inhibit both soluble and fibrin-bound thrombin and do not bind to plasma proteins. They also do not affect platelets. These effects provide significant advantages over heparin.

The first synthetic thrombin active site inhibitor, argatroban, was introduced, followed by the discovery of many other direct thrombin inhibitors. Most of these inhibitors are peptidomimetic compounds that bind to the active site of thrombin. Bivalirudin, argatroban, lepirudin, and dabigatran are among the DTIs that have been developed and are in advanced use [20].

• Bivalirudin

Bivalirudin is a direct thrombin inhibitor that binds directly and reversibly to thrombin, inhibiting both circulating and clot-bound thrombin. These effects are advantageous compared to heparin. In adults, its duration of action is approximately 25 minutes. It has been shown to have stable pharmacokinetics, a rapid anticoagulant effect, easy dose titration, and no risks for thrombocytopenia [21]. It does not require antithrombin for its effect and does not inhibit Factor Xa. Bivalirudin is metabolized by proteolytic enzymes and approximately 20% is excreted by the kidneys. There is no antidote available. However, a potential disadvantage is its ability to promote clotting in areas of low or stagnant flow. Common side effects include hypotension, back pain, and nausea, while rare side effects include bleeding, coronary artery stent thrombosis, ventricular fibrillation, and kidney failure. The indications for bivalirudin are shown in Table 1.

FDA-labeled indications	Percutaneous coronary intervention; Heparin-induced thrombocytopenia
Off-label indications according to the FDA	Acute myocardial infarction; Percutaneous coronary intervention
	Heparin-induced thrombocytopenia with thrombosis and cardiac surgery
	Deep vein thrombosis
	Peripheral vascular surgery
	Thromboembolic disorders
	Unstable angina
	Adjunct to thrombolytic therapy

Table 1.

Bivalirudin indications.

In our cardiac surgery center, the use of bivalirudin during ECMO and ECMO management.

The utilization of bivalirudin is relatively recent compared to heparin and is potentially applicable as an anticoagulant during ECMO procedures. Bivalirudin is recommended as the second-line anticoagulant in the ELSO guidelines (2021). However, the use of bivalirudin during ECMO in ECMO centers is only about 6% [22]. The initial and maintenance doses of bivalirudin, as well as the anticoagulation parameters and target values to be monitored during ECMO, have not been definitively established yet. There are limited institutional protocols in place [23]. Due to the lack of international consensus on the dosage and titration of anticoagulant drugs, it is necessary to establish institutional protocols. The therapeutic response for patients receiving bivalirudin is most commonly assessed using activated aPTT, with ACT measurements often used in conjunction. In adult patients, the average (IQR) daily bivalirudin doses required are higher in COVID-19-related ARDS compared to nonCOVID-19-related ARDS (3.1 μg/kg/min vs. 2.4 μg/kg/min) [18]. The implementation of institutional bivalirudin titration nomograms increases the number of aPTT values within the therapeutic range (from 48 to 62%). The daily titration frequency decreases to achieve aPTT within the therapeutic range [24]. Bivalirudin is administered at a dose of 0.15–0.25 mg/kg/hour when targeting APTT values between 60 and 80 seconds [24, 25, 26]. The time to reach the therapeutic range when monitored using APTT is approximately 20 hours [26].

In patients undergoing continuous renal replacement therapy (CRRT) during ECMO, the dose of bivalirudin is lower compared to patients not receiving CRRT (0.15 ± 0.06 mg/kg/hour vs. 0.28 ± 0.36 mg/kg/hour) [22]. In neonates and small children, antithrombin levels are low, and due to the immaturity of the liver, hemostatic proteins are insufficient. Therefore, anticoagulation management during pediatric ECMO is more challenging. In pediatric ECMO applications, there is no correlation between the dose of bivalirudin or heparin and laboratory parameters such as prothrombin time, thromboelastography, and antifactor Xa. There is also no significant relationship between bivalirudin dose and thromboelastographic parameters like TEG-R or aPTT [27]. However, the presence of a correlation between bivalirudin dose and PTT, INR, and ACT has been reported [28]. Particularly in the pediatric patient group, multimodal anticoagulation monitoring is crucial [23]. Compared to heparin, the risk of thrombosis in both the device and the patient is lower with bivalirudin [29, 30, 31]. New venous thromboembolic events with bivalirudin are reported to be 6.1% [26]. There are similar bleeding rates during ECMO with heparin or bivalirudin [32]. A meta-analysis comparing heparin (UFH), argatroban, bivalirudin, and/or nafamostat found no significant difference in major bleeding risk between the groups [33]. The bleeding rates during ECMO with bivalirudin range from 15.1 to 28% [26, 34]. Bivalirudin is associated with a reduced need for packed red blood cells and platelet transfusions (means of 13 vs. 39, p = 0.004; 5 vs. 19, p = .014, respectively) [29].

In patients treated with bivalirudin during ECMO, the average circuit lifespan is longer compared to heparin (16 vs. 10 days, p = 0.004) [29]. There is a significant increase in the time to circuit thrombosis (p = 0.007) [28]. Circuit change due to clotting or inadequate oxygenation occurs in nonCOVID-19 patients at a rate of 36% [34]. Compared to UFH and argatroban, bivalirudin reduces the risk of device-related and patient-related thrombosis. It appears to be a better choice for patients undergoing ECMO [31]. Short-term mortality is lower with bivalirudin compared to heparin in patients undergoing ECMO [35].

ELSO member center for cardiac surgery, since the onset of the Covid-19 pandemic, bivalirudin has been used in ECMO applications for patients with ARDS due to known deficiencies of heparin as an anticoagulant. Currently, bivalirudin is preferred as the anticoagulant in all ECMO applications. In our center, during the pandemic period, in 52 patients who underwent V-V ECMO due to ARDS related to Covid-19, bivalirudin was used at a dose of 0.03–0.04 mg/kg/hour initially, with an average dose of 0.036 mg/kg/hour during the early stages of ECMO. The target APTT value was set at 45–60 seconds, and ACT was targeted at 160–200 seconds. Bivalirudin dosage adjustments were made by monitoring APTT and ACT concurrently. Platelet counts were monitored daily, and D-dimer and lactate dehydrogenase levels were monitored every few days. Dose adjustments were made by an experienced and dedicated team familiar with ECMO. Arterial blood gas analysis was performed twice daily for the assessment of ECMO oxygenator performance, both before and after the oxygenator, and when necessary, the fresh gas flow was increased to 10 liters per minute for a few minutes. Washing of the ECMO oxygenator membrane with up to 300 cc of crystalloid fluid was performed when deemed necessary. This meticulous anticoagulation and mechanical methods ensured the preservation of oxygenator and circuit performance. A successful ECMO management with bivalirudin infusion and a 71-day vv-ECMO duration, followed by uncomplicated discharge, was reported in a patient with Covid-19-associated ARDS treated using these methods [33]. It is important that both anticoagulation management and overall patient monitoring be performed by a multidisciplinary team experienced in ECMO, 24 hours a day, 7 days a week, and utilizing all available technological resources [36].

• Argatroban

Argatroban is a DTI that reversibly binds to the active site of thrombin, recommended as one of the anticoagulants for use during ECMO. It has a half-life of 39–51 minutes, does not require antithrombin, lacks an antidote, and has varying dose requirements among individuals. It is metabolized in the liver and excreted via feces [28]. It is FDA-approved for the prophylaxis and treatment of thrombosis in patients with HIT and heparin-induced thrombocytopenia and thrombosis syndrome.

The use of direct thrombin inhibitors such as argatroban is recommended in suspected or confirmed cases of HIT, HIT-like conditions, sepsis or COVID-19 patients, patients with extracorporeal circuits, and those with heparin resistance [37]. Argatroban exhibits similar efficacy and safety to lepirudin and bivalirudin in HIT patients [38]. One significant side effect is serious neutropenia, which may require the administration of granulocyte colony-stimulating factors upon discontinuation of argatroban [39]. Although active PTT monitoring is commonly used for argatroban, diluted thrombin time (TT) may provide more precise measurements, especially in critically ill patients. Monitoring with bedside viscoelastic point-of-care testing guides in detecting overdosing [40].

The disadvantages associated with the use of DTIs during ECMO include limited availability of specific laboratory monitoring capabilities, lack of a specific antidote, and limited experience with ECMO. Prospective randomized studies are needed to confirm the effectiveness and superiority of DTIs as primary anticoagulants in ECMO patients [21].

• Lepirudin

Lepirudin is a direct thrombin inhibitor that inhibits platelet aggregation and thrombus formation. It does not cross-react with HIT antibodies and lacks an antidote. Its half-life increases in patients with renal failure. Monitoring of lepirudin anticoagulation can be achieved using aPTT and ACT. Apart from successful case presentations related to lepirudin use, there are no retrospective or prospective controlled studies available.

2.3.3 Nafamostat

Nafamostat is a synthetic serine protease inhibitor. It inhibits thrombin, factor Xa, XIIa, the kallikrein-kinin system, the complement system, and nitric oxide production. Its half-life is 8–10 minutes. In cases of repeated clotting of CRRT filters in patients at high risk of bleeding, alternative anticoagulation with nafamostat is recommended. Baek et al. applied systemic anticoagulation with NM to critically ill patients at high risk of bleeding during ECMO [41]. Lang et al. noted a limited number of studies with control groups in this regard [42]. Prospective studies are needed to determine the optimal use of this approach during ECMO [43].

2.4 Monitoring anticoagulation during ECMO

Achieving optimal anticoagulation during ECMO without causing bleeding or thrombosis is quite challenging. There is no specific test to assess this. Factors such as the patient’s age, underlying disease, organ dysfunctions, thrombotic or bleeding diathesis, and concomitant medications are important. The type and dosage of the anticoagulant drug should be carefully monitored along with coagulation tests that can monitor the anticoagulant drug. Depending on the capabilities and resources of the centers, ACT, activated aPTT, antithrombin (AT), antifactor Xa (anti-Xa), international normalized ratio (INR), platelet count, and bedside viscoelastic tests (thromboelastography (TEG®), rotational thromboelastometry (ROTEM)) should be evaluated alone or in combination.

ACT is preferred to rapidly adjust the anticoagulant dose upon initiation of ECMO. It measures the time to initial fibrin formation. However, various factors such as hemodilution, hypothermia, platelet function and count, hypofibrinogenemia, and deficiencies in coagulation factors can influence ACT [44]. Therefore, in many centers, anticoagulation monitoring is performed using a combination of ACT and various other tests.

The target ACT level is generally set at 200 seconds for VA-ECMO and 160–180 seconds for VV-ECMO, adjusted according to the patient’s coagulation parameters. Heparin is typically administered as a bolus during ECMO application, and both ACT and APTT are evaluated every hour; this frequency is reduced to approximately every 4–6 hours once the patient’s condition stabilizes.

In 2019, in 108 neonatal and pediatric ECMO centers registered with the Extracorporeal Life Support Organization in the United States, antifactor Xa and activated aPTT combination (%68) or antifactor Xa and aPTT combination (%54) were reported to be used for monitoring heparin therapy. Antifactor Xa levels were utilized in 90% of cases, while viscoelastic tests were employed in 41% of cases to assist in anticoagulation management. Compared to 2013 results, heparin remains the most frequently used anticoagulant, and antifactor Xa test combination is preferred for monitoring [21].

The aPTT test measures clot formation in plasma optically or mechanically. During ECMO, values of 60–80 seconds are commonly used, with a target of 40–60 seconds for patients at high risk of bleeding [25, 40, 41]. The activated aPTT is the most common test used for anticoagulation monitoring during ECMO (86% of ECMO days), with an average of 52 seconds (range 39–61). However, aPTT is 5 seconds lower after the first bleeding episode [6]. Additionally, monitoring antithrombin (AT) levels is recommended alongside aPTT and antifactor Xa for appropriate anticoagulant effect monitoring [6].

In young children, the physiological concentrations of natural anticoagulants are typically low. Interpreting neonatal coagulopathy is challenging, and there are gaps in knowledge regarding anticoagulation in newborns. Especially, monitoring antithrombin (AT) levels and its necessity in neonates are still debated. The target antifactor Xa value set for therapeutic anticoagulation is 0.3–0.7 IU/mL [45]. However, there is no correlation between anti-Xa levels and ACT values. Anti-Xa activity may be a more suitable test for monitoring anticoagulation [46]. In hospitalized patients, those monitored with anti-Xa have lower transfusion rates compared to those monitored with aPTT [47].

A multifaceted approach to anticoagulation monitoring during ECMO (utilizing antifactor Xa tests, thromboelastography, and antithrombin measurements together) has been associated with a reduction in blood product transfusions and hemorrhagic complications, as well as an increase in circuit lifespan [48].

During ECMO support, daily monitoring of laboratory values including platelets, hemoglobin, antithrombin III, and fibrinogen is necessary due to their potential impact on coagulation. Hemolysis may occur during ECMO, therefore daily monitoring should include bilirubin, hemoglobin, haptoglobin, and lactate dehydrogenase [48].

Viscoelastic tests (ROTEM or TEG) allow for real-time assessment of clot formation rate, strength, and clot lysis [49]. For evaluation of platelet function disorders or von Willebrand factor deficiency, it is more suitable to use modified TEG Platelet Mapping (TEG-PM) and ROTEM with a multiple-electrode aggregometry analyzer. Studies using viscoelastic tests for the evaluation of coagulopathy and monitoring of anticoagulation are needed in both pediatric and adult populations [50].

There are few institutional protocols for monitoring anticoagulation [32]. Currently, there is no international consensus on the specific tests or target laboratory values to be used for monitoring anticoagulation.

2.5 Evaluation of bleeding and thrombosis during ECMO

Standardizing the definitions and management of bleeding and thrombosis during ECMO is crucial for conducting more meaningful studies and achieving more precise results. Therefore, a group of experts in cardiac surgery and mechanical circulatory support, convened by the National Heart, Lung, and Blood Institute and the US Department of Defense, aimed to establish global consensus definitions for evaluating hemostatic outcomes more meaningfully. To this end, consensus recommendations for primary outcomes were developed. Recommendations for hemostasis were determined under the title “Ordinal Composite Efficacy Endpoint Based on Common Terminology Criteria for Adverse Events (Version 5.0)”. Factors such as quantities of packed red blood cell use, blood product utilization, hemodynamic compromise, and others were included for bleeding or thrombosis assessment. Evaluations for thrombosis included increases in transmembrane pressure, arterial–venous thrombosis, ECMO circuit changes, thrombectomy, among others [51].

There is a current trend toward a new approach in managing coagulation disorders, aiming to enhance the effectiveness and safety of anticoagulation by integrating traditional or newer anticoagulant drugs with modern technological developments. In achieving optimal anticoagulation during ECMO, there is a tendency toward a multidisciplinary approach that utilizes pharmacokinetics and personalized dosage algorithms. The management of anticoagulation should be individualized for each patient, taking into consideration the patient’s characteristics, clinical condition, and laboratory results. Treatment plans are adjusted based on the individual’s genetic predisposition to clotting and response to anticoagulants, aiming to minimize adverse effects and achieve optimal outcomes. Continuous evaluation of the clotting status ensures timely and appropriate anticoagulation therapy. The implementation of the latest technologies such as artificial intelligence and machine learning may enhance the safety of anticoagulation.

Advancements in anticoagulant therapy continue to progress in the field of ECMO. Utilizing real-time monitoring and advanced technologies, the focus is on implementing individually tailored treatment strategies for anticoagulant therapy and managing coagulation disorders.

2.6 ECMO circuits and anticoagulant coating

Heparin coating is widely used in ECMO circuits today. To prevent ECMO thrombosis without the use of systemic anticoagulant drugs, coating ECMO membranes with anticoagulant drugs is being considered. Hydrogel coating with bivalirudin, which is still in the experimental stage, does not affect coagulation levels and reduces the risk of complications such as systemic bleeding compared to intravenous injection [52]. Phosphorylcholine/heparin composite and polycarboxybetaine coatings are still in the experimental stage [53].

3. Conclusion

In conclusion, the use of anticoagulant medication during ECMO is essential to sustain ECMO support. However, the selection of anticoagulant medication and management of anticoagulation should be performed by an experienced ECMO team. In the future, the goal may be the development of biocompatible ECMO oxygenators and circuits that do not require systemic anticoagulation during ECMO support. Such advancements could contribute to making ECMO therapy safer and more effective by reducing or eliminating the use of anticoagulant medications, thereby minimizing potential adverse effects and improving the treatment process. This could represent a significant advancement in both patient outcomes and treatment costs.