Results
Between September 2020 and December 2022, a total of 19 patients (Table 1) were identified as having received apixaban at the time of transplantation (42% male, median 13.5 years of age). Of those, 13 patients had congenital heart disease, with the remainder having cardiomyopathy. Sixteen of the 19 patients were listed United Network for Organ Sharing (UNOS) Status 1A, of which 3 patients were supported with a HeartMate 3 (HM3, Abbott, Chicago, IL), 2 patients were Status 1B, and 1 patient was Status 2. Fifteen patients were maintained on inotropic support with milrinone infusions in inpatient and outpatient settings. For 13 patients, apixaban was administered in 3 patients for prophylaxis due to history of previous thrombus, 4 patients with severe ventricular dysfunction with ejection fraction <30%, and 6 patients for the presence of a central venous line in single ventricle palliation. Therapeutic levels were targeted in 6 patients for active thrombus treatment or presence of VAD. All patients had normal kidney function at the time of transplant based on serum blood urea nitrogen and creatinine levels.
Median length of apixaban therapy prior to transplantation was 114.6 days (Table 2). Dosing ranged from 0.625mg twice daily to 7.5mg twice daily. In addition, the 3 patients on HM3 VAD and 1 patient with a bioprosthetic mitral valve received concomitant antiplatelet therapy with aspirin (ASA). There were no clinically relevant bleeding, including no major bleeding, or thrombotic events while awaiting transplant. The median time from last apixaban dose to arrival to the OR was 23.2 hours. The median apixaban level prior to OR arrival was 37ng/mL, obtained 6.2 hours prior to OR arrival. Apixaban level testing was not obtained immediately prior to transfer to the OR in 3 patients. The median chest tube output in the CICU post-operatively was 1.9mL/kg/hr and 1.8mL/kg/hr for the first 2 and 4 post-operative hours respectively.
In total, 6 patients met our definition for significant post-operative bleeding, including 1 patient for chest tube output> 10mL/kg/hour for the first two hours, 4 patients for chest tube output > 5mL/kg/hour over the first 4 hours, and 1 patient requiring re-intervention for hemodynamically significant bleeding. No patients developed tamponade. All 6 of these patients had palliated complex congenital heart disease (Table 3) with a median of 3.5 sternotomies per patient (3, 4.8) and median age of 6.1 years (3.6, 13.1). Additionally, 3 patients had anti-Xa levels post-operatively <0.2 units/ml, an average of 10 hours after arrival to the intensive care unit. This confirms no measurable residual effect of apixaban.
A number of adverse postoperative events deserve further description. As seen in Table 3, Patient 6 developed severe primary graft dysfunction requiring extracorporeal membrane oxygenation (ECMO) in the first 12 postoperative hours. This patient was the only to receive a reversal agent recombinant Factor Xa (andexanet alfa) in addition to protamine, despite no notable increase in operative bleeding or difficulties achieving hemostasis. Reason for administration of recombinant Factor Xa was that it was thought to help establish hemostasis in a patient with previous sternotomies and it was made readily available for administration. Notably, pre-operative apixaban level was low at 28.4ng/ml. While on ECMO, patient underwent cardiac catheterization demonstrating multiple diffuse pulmonary embolism. This patient developed gross hemolysis, with multi-organ failure and subsequent redirection of care and death on post-operative day 7. Patient 12 required re-exploration for bleeding after development of acute postoperative thrombosis in the reconstructed systemic venous system requiring trans-catheter stent placement near a fresh suture line. It was thought at the time the significant bleeding was mechanical in origin post-stent placement, not related to anticoagulation. Patient 14 was managed on HM3 ventricular assist device with apixaban and ASA prophylaxis pre-transplant without significant complications. Postoperatively, he was noted to have severe global neurologic injury unrelated to bleeding or thrombosis and not left to be attributable to apixaban, resulting in death.