Recombinant factor VIIa while on extracorporeal membrane oxygenator support: a word of caution †

TLDR: It is shown that patients receiving recombinant factor VIIa (rFVIIa) for intractable bleeding after cardiac surgery are not at an increased risk of thromboembolic events, and this effective haemostatic agent can be used with an acceptable safety profile in this patient population.

Abstract: We read with interest the article by Chapman et al. [1]. They showed that patients receiving recombinant factor VIIa (rFVIIa) for intractable bleeding after cardiac surgery are not at an increased risk of thromboembolic events. Therefore, they concluded that this effective haemostatic agent can be used with an acceptable safety profile in this patient population. We would like to ask the authors whether, among the 236 patients receiving rFVIIa, they had patients supported by extracorporeal membrane oxygenator (ECMO). Bleeding is a major problem in postoperative ECMO implantation for post-cardiotomy cardiogenic shock [2]. Severe haemorrhage requiring re-exploration occurs in up to 58% of cases [3] and carries a dismal prognosis. Among patients who returned to the operating theatre, a surgical source of bleeding is not identified in more than half of the patients [1]. Hence, the off-label use of rFVIIa represents an attractive haemostatic agent for controlling bleeding attributed to a disseminated intravascular coagulation-like phenomenon [4]. We entirely agree with their opinion about the effectiveness of rFVIIa as a haemostatic agent, and we would like to report an exceptional thrombotic complication that we recently encountered: a native non-calcified aortic valve thrombosis in a patient on ECMO support who received rFVIIa for massive bleeding after coronary artery bypass grafting (CABG) surgery. A 60-year old male patient was admitted on an elective basis for CABG surgery. His past medical history included hypertension and multiple sclerosis. Coronary angiogram revealed triple vessel disease. Transthoracic echocardiography disclosed a normal aortic valve and an ejection fraction at 45%. The patient had a normal hepatic and renal function and a normal coagulation profile before surgery. Cardiopulmonary bypass (CPB) was established through median sternotomy. The obtuse marginal artery was not identified during surgery due to severe adherence between the lateral aspect of the left ventricle and the pericardium while the right coronary artery and the left anterior descending artery were bypassed. Weaning from CPB was unsuccessful, and femoro-femoral ECMO was instituted. The patient was transported to the angiography laboratory where a drug eluting stent was successfully implanted in the proximal circumflex artery. Massive bleeding from the chest tube was recorded. Therefore, a 90-μg/kg of rFVIIa was infused, and bleeding decreased. Heparin infusion was commenced 18 h after the surgery. On postoperative day (POD) 1, a transoesophageal echocardiogram (TEE) revealed severely depressed contractility of the left ventricle and thrombus formation on the three cusps of the aortic valve on the aortic side. Despite full heparinization, a complete thrombosis of the aortic valve was disclosed on POD 3, and no clots were noted in the ECMO tubing. His family refused permission for further treatment and he died after the removal of the support. In view of our experience with this drug, we recommend caution when rFVIIa is used in the postoperative period after cardiac surgery in the setting of ECMO support; careful patient management by routinely performing TEE and maintaining inotropic support for ventricular contractility and cusps mobility is mandatory to prevent thrombus formation on the aortic cusps.