Andexanet Alfa in Urgent Cardiac Surgery: A Case Report of Edoxaban Reversal for Acute Hemopericardium

Introduction

Non-vitamin K oral anticoagulants are frequently prescribed to patients with thromboembolic disease or nonvalvular atrial fibrillation to prevent systemic embolism [1,2]. The major advantage of these substances is that they do not require routine hemostatic monitoring, unlike warfarin derivates [3]. Recently, antidotes like idarucizumab (Praxbind) and andexanet alfa (Ondexxya) have become available for reversing the anticoagulant effects of dabigatran (Pradaxa) and apixaban (Eliquis)/rivaroxaban (Xarelto), respectively [4,5].

Andexanet alfa (Ondexxya, AstraZeneca GmbH) is an inactive, recombinant, modified human factor Xa decoy protein that binds to factor Xa inhibitors and reverses their activity [5]. The Food and Drug Administration (FDA) and the European Medicines Agency (EMA) conditionally approved andexanet alfa for reversing anti-factor Xa activity in life-threatening or uncontrolled bleeding in patients on apixaban or rivaroxaban [6]. Although edoxaban is also an oral factor Xa inhibitor, andexanet alfa has not been approved for reversing its anticoagulant effects due to insufficient data [4,6].

The Society of Thoracic Surgeons (STS)/Society of Cardiovascular Anesthesiologists (SCA)/American Society of ExtraCorporeal Technology (AmSECT)/Society for the Advancement of Blood Management (SABM) guidelines on patient blood management recommend using specific reversal agents – idarucizumab for dabigatran and andexanet alfa for apixaban or rivaroxaban – in emergency cardiac surgery [7]. If the appropriate antidote is unavailable, prothrombin complex concentrate is recommended as an alternative [7]. Notably, the use of andexanet alfa to reverse edoxaban before cardiac surgery is not supported by current guidelines [7].

This study evaluated the off-label use of andexanet alfa for reversing edoxaban in the management of acute hemopericardium requiring urgent cardiac surgery. It highlights the practical challenges and discusses the clinical outcomes of addressing life-threatening bleeding scenarios, particularly with the use of andexanet alfa.

Case Report

A 71-year-old woman with a history of hypertension, urothelial carcinoma in remission, and persistent atrial fibrillation (AF) presented to the hospital with severe dyspnea and pulmonary edema resulting from a hypertensive crisis and atrial flutter. Her long-term treatment included edoxaban 60 mg daily for anticoagulation.

Upon admission to the intensive care unit (ICU), the patient received intravenous antihypertensive therapy (urapidil 25 mg) and diuretics (furosemide 40 mg). Emergency electrical cardioversion successfully restored sinus rhythm. Transthoracic and subsequent transesophageal echocardiography revealed normal left ventricular ejection fraction but severe mitral valve insufficiency (grade II–III). To evaluate the mitral insufficiency and rule out pulmonary hypertension or coronary artery disease (CAD), the team planned a right-heart catheterization and coronary angiography.

On the day of the procedure, edoxaban was withheld. After placement of a 5-French sheath in the right radial artery and a 7-French sheath in the right femoral vein, the patient received 5000 units of unfractionated heparin (UFH) and 0.3 mg of nitroglycerin to prevent radial artery occlusion. A 7-French pulmonary artery catheter (Corodyn P1) was advanced for right-heart catheterization.

During the procedure, the patient developed severe arterial hypotension (blood pressure 70/40 mmHg), agitation, confusion, and rapidly altered consciousness. Jugular vein distension and cyanosis were noted, indicating acute cardiac tamponade. Transthoracic echocardiography confirmed the diagnosis (Figure 1).

An emergency pericardiocentesis was performed via a subxiphoid approach, and a pigtail catheter was placed for continuous drainage (Figure 2). Despite aspiration of 500 mL of blood, there was no clinical improvement or resolution of the pericardial effusion on echocardiography. The patient’s hemodynamic instability necessitated intubation, mechanical ventilation, and catecholamine support.

Due to ongoing bleeding and lack of hemostasis, 5000 units of protamine (to reverse UFH) and 400 mg of andexanet alfa (to reverse edoxaban) were administered intravenously. Following this, the pericardial effusion regressed under repeated suction, and hemodynamic stabilization was achieved after draining a total of 1100 mL of blood (Figure 3). The patient received 3 units of red blood cell concentrates and 1 liter of cell-salvaged blood. Norepinephrine infusion was discontinued after effusion resolution.

The patient was transferred for emergency cardiac surgery. Preoperative anti-factor Xa activity exceeded 1.5 IU/mL, as shown in Table 1. During surgery, a perforation at the apex of the right ventricle was identified and sutured. Intraoperative transesophageal echocardiography confirmed a severe mitral valve insufficiency. Given the emergency situation and the functional nature of mitral valve dysfunction, a mitral valve replacement was performed using a 31-mm pericardial prosthesis (Perimount Magna Ease).

For cardiopulmonary bypass (CPB), 40 000 units of UFH and 1000 units of antithrombin III were required to achieve an activated clotting time (ACT) of 600 seconds. CPB lasted 67 minutes, with 44 minutes of aortic clamping. Protamine (28 000 IU) successfully reversed heparin after CPB, and no clotting factors were required intraoperatively or postoperatively. The patient’s hemoglobin was 11.4 g/dL after transfusion of 2000 mL of cell-salvaged blood.

The patient was transferred to the ICU with stable hemodynamics and mechanical ventilation. Extubation occurred on postoperative day 1, with no neurological deficits. Renal function remained consistent with preoperative levels of moderate impairment. During the hospital stay, the patient received 5 units of red blood cell concentrates (3 in the catheter lab and 2 postoperatively), 2 fresh frozen plasma units, and 2 platelet concentrates.

Due to atrial fibrillation, she was cardioverted into sinus rhythm. No excessive bleeding or thrombotic complications occurred postoperatively. A detailed timeline of events is presented in Table 2.

Discussion

Through the management of our 71-year-old patient on edoxaban therapy, who experienced an unexpected, rare complication of acute hemopericardium during diagnostic right-heart catheterization, we gained valuable insights into handling life-threatening bleeding requiring urgent cardiac surgery. This case highlights the critical importance of promptly reversing anticoagulation with agents like andexanet alfa, while also underscoring the need to carefully weigh the associated risks. Furthermore, we learned about the unique challenges of heparinization before cardiac surgery and the nuanced role of these reversal agents in such scenarios. Lastly, the case emphasized the necessity of seamless collaboration among various specialists, including cardiologists, anesthesiologists, and surgeons, to manage such complex situations effectively.

The exact etiology of the perforation was not clear, but it is suspected to be the use of J-shaped wire to support the pulmonary catheter along with a dilated thin-walled right ventricle. Nevertheless, swift intervention through pericardiocentesis, along with antagonization of both heparin and edoxaban, was pivotal to stabilizing the patient in the cardiac catheter laboratory, especially as a safe bridge to cardiac surgical intervention. Notably, andexanet alfa was utilized off-label in this instance. The decision was guided by an assessment of andexanet alfa’s mechanism of action and the urgent, life-threatening nature of the situation.

Andexanet alfa was conditionally approved by the FDA and EMA for the reversal of anti-factor Xa activity in cases of life-threatening or uncontrolled bleeding associated with apixaban and rivaroxaban [6]. This conditional approval does not include edoxaban or perioperative risk mitigation. Although some earlier reports documented the safe and effective administration of andexanet alfa to reduce perioperative risk [8–10], the associated thromboembolic risk restricts its broader use [11]. The ANNEXA-I trial demonstrated a significant thromboembolic risk associated with andexanet alfa, as thrombotic events occurred in 10.3% of patients treated with andexanet, compared to 5.6% in those receiving usual care. Additionally, ischemic stroke was observed in 6.5% of the andexanet group, compared to 1.5% in the usual care group [11]. This risk is attributed to both the underlying conditions necessitating anticoagulation and the pro-coagulant effect of andexanet alfa, partly due to its inhibition of tissue factor pathway inhibitor (TFPI) [4].

During heparinization prior to cannulation for cardiopulmonary bypass (CPB), we were not able to achieve an adequate ACT until we administered of 40 000 units of heparin and 1000 units of antithrombin III. Many reports observed altered heparin responsiveness or heparin resistance [9,12] after andexanet administration. This led EMA to issue a document advising against the use of andexanet alfa prior to cardiac surgery [13]. This further highlights the questionable safety of andexanet alfa in the context of cardiac surgery

The approval did not include edoxaban, as the data were insufficient for approval. The phase III ANNEXA-I trial did not include patients treated with edoxaban [6]. Moreover, the anti-factor Xa activity decreased in patients with edoxaban (reduction 71%) less than with apixaban (93%) or rivaroxaban (94%) [4]. On the other hand, the percent of patients with good or excellent hemostasis after andexanet alfa in patients on edoxaban was 79% (95% C.I. 59–92%), which was comparable to patients on rivaroxaban 81% (95% C.I. 73–87%) and apixaban 79% (95% C.I. 72–85%) [3]. The positive hemostatic effect of andexanet alfa may not only be mediated through antagonizing anti-factor Xa activity. As mentioned above, andexanet alfa inhibits tissue factor pathway inhibitor. Moreover, other effects may be mediated through endogenous thrombin potential (ETP), as the ANNEXA-I trial showed that ETP returned to the normal range by the end of andexanet alfa bolus for 24 hours in all FXa inhibitors [3,11]. Adexanet alfa has been approved in Japan to reverse edoxaban in life-threatening bleedings [14].

In a broader context, anti-factor Xa assays calibrated with reference ranges intended for unfractionated heparin or low-molecular-weight heparin can serve as valuable tools for estimating the relative anticoagulant activity present in a patient. However, the correlation with hemostatic efficiency remains uncertain.

Although in our patient the last dose of edoxaban was given after more than 24 hours before andexanet alfa was given, the anti-factor Xa level was >1.5 IU/mL at least 2 hours after administering the andexanet alfa bolus, indicating persistent anticoagulant activity. Because the half-life of edoxaban is 10–14 hours [15], it is unlikely for edoxaban to be in therapeutic range after this period. Heparin probably contributed to this high level of anti-factor Xa activity, even though heparin was also reversed with protamine. Furthermore, studies have shown that significant dilution occurs during blood sample preparation, causing dissociation of the andexanet-direct anticoagulant complex and leading to an artificial increase in anti-FXa activity [14]. Therefore, measuring anti-factor Xa activity after administering andexanet alfa is not recommended [14]. Notably, the tests used in ANNEXA-I and ANNEXA-4 were modified to minimize dilution in test systems, but these tests are not commercially available [14].

Another important point is the significant interactions between andexanet alfa and other coagulation tests, such as partial thromboplastin time (PTT), which makes managing unfractionated heparin (UFH) during procedures like cardiac surgery challenging [14]. Our patient only received the bolus injection of andexanet alfa, which likely was a dosing error.

4-factor prothrombin complex concentrate (4F-PCC) is a combination of blood clotting factors II, VII, IX, and X, which is typically used to reverse vitamin K antagonists. Nevertheless, this complex has been used to achieve hemostasis in patients with major bleeding who are receiving anti-factor Xa inhibitors [16]. Some observational studies have compared 4F-PCC with andexanet alfa in anti-factor Xa activity reversal, yielding conflicting results [17,18].

The cost of anti-factor Xa inhibitors antidotes is an important aspect of clinical practice, and this has been addressed in several studies [19,20]. While this may seem important at first glance, it becomes justifiable when viewed in the context of potential complications that could arise in the absence of antidotes.

Conclusions

We report an unconventional application of andexanet alfa, which contributed to stabilizing the patient without apparent adverse effects. However, it highlights the complex challenges involved in reversing anticoagulation, particularly for agents like edoxaban that lack an approved antidote. The known increased thromboembolic risks associated with andexanet alfa, along with the development of heparin resistance, necessitate careful consideration of potential risks prior to its use, especially in the context of cardiac surgery.

The successful management of this case was achieved through a multidisciplinary approach with close collaboration between cardiologists, cardiac surgeons, and anesthesiologists. It is essential that further research evaluates the safety and efficacy of andexanet alfa in managing major bleeding in patients on edoxaban therapy, providing clinicians with more definitive guidance in these complex scenarios.