Stratmann G, Russell IA, Merrick SH:, Use of recombinant factor VIIa as a rescue treatment for intractable bleeding following repeat aortic arch repair Ann Thorac Surg 2003;76:2094-2097.
The authors report a case of life threatening hemorrhage, refractory to aggressive conventional therapy, at a rate of 16 L/hr following cardiac surgery and aortic arch repair, that was controlled with a dose of 90 µg/kg of recombinant factor VIIa (rFVIIa), repeated once after 2 hours. A 60-year-old man had undergone a Bentall aortic root replacement 19 years previously using a porcine-valved conduit for acute type A aortic dissection, and scheduled for elective aortic valve replacement for severe prosthetic valve insufficiency and aneurysmal enlargement of the remaining ascending aorta. The patient was recovering from a lobar pneumonia for which he was taking ciprofloxacin by mouth and was not receiving any anticoagulants.
Intraoperatively, the patient received aprotinin, 2 x 106 kallikrein inhibitory units (KIU) as an initial load followed by 500,000 KIU/h for 6 hours. Bleeding from an accidental aortotomy upon sternotomy required rapid transfusion of 8 U of packed red blood cells (PRBC) and reinfusion of 550 ml of shed blood while CPB was being instituted emergently reducing the patient's core temperature to 20°C.. Anticoagulation with heparin 400 U/kg produced an activated clotting time of >600 seconds and a heparin concentration of 4.5 mg/kg. A vertical laceration was found in the native, dissected ascending aorta. The previous Bentall aortic root replacement covered only a 3-cm segment of the proximal ascending aorta. The remainder of the aneurysmal ascending aorta and proximal aortic arch was mobilized during the cooling period. At onset of ventricular fibrillation, an aortic cross-clamp was applied. The Dacron valved conduit was then opened and cardioplegia administered directly into the coronary ostia. A degenerated porcine valve was then removed from the base of the Dacron conduit and a St. Jude aortic prosthesis reinserted. At this point, circulation was arrested and the aortic cross clamp was removed. The dissected ascending aorta and proximal arch were removed and the remaining dissection flap was fenestrated. A new Dacron graft was interposed between the arch and original proximal graft. No bioglue or Teflon felt was used, as the residual distal aorta and proximal aortic arch were fibrotic from chronic dissection. Total circulatory arrest time was 20 minutes.
Separation from CPB was complicated by profuse bleeding from the multiple suture sites, which was difficult to control, requiring reinitiation of CPB three times in order to facilitate exposure of bleeding sites on the pulmonary artery and posterior aortic suture line. Final separation from CPB (total CPB time of almost 6 hours) was accompanied by diffuse bleeding and coagulopathy. Inotropic/vasoactive support with dopamine (5 µg/kg/min), norepinephrine (0.05 µg/kg/min) and phenylephrine at varying doses and vigorous replacement of rapid blood loss prevented the imminent cardiovascular collapse. After protamine reversal of heparin, blood loss persisted at a rate of 270 mL/min despite a heparin concentration of zero mg/kg. Over the next 90 minutes, 25 U of PRBC (6,250 mL), 22 U of fresh frozen plasma (FFP) (4,400 mL), 10 six-packs of platelets (Plt) (2,000 mL), and 8 ten-packs of cryoprecipitate (800 mL), in addition to 11,000 mL of shed mediastinal blood, were administered with no apparent therapeutic effect. The authors estimated that the replacement of the patient's blood volume 4 to 6 times likely reduced the concentrations of some critical coagulation factors including activated factor VII to levels too low to support thrombin formation.
Based on the life threatening hemorrhage, a dose of 90 µg/kg of recombinant factor VIIa (rFVIIa) was administered intravenously. Three minutes later the bleeding had slowed down, so that rapid resuscitation with blood products was no longer necessary. The patient received no more blood products except for six units of platelet concentrate. An empiric second dose of rFVIIa was administered 2 hours after the initial dose had been given. This was based on therapy in hemophiliacs, and due to the plasma halflife of rFVIIa of 2.3 to 2.9 hours. The patient recovered slowly, was extubated on the third postoperative day, and left the hospital six days later without any symptoms or signs of systemic thromboses.
The authors suggest that rFVIIa should currently only be used if all reasonable surgical attempts have been made to control bleeding and if a coagulopathy is uncorrected despite aggressive administration of standard blood products. The authors conclude that despite all concerns, the lifesaving potential of rFVIIa must not be underestimated.