rFVIIa – Here’s a heated (but professional)
exchange regarding the use of NovoSeven
in CABG postoperative hemorrhage. This
correspondence is from the Annals of
Thoracic Surgery (2007) 83:354–60.
ncbi.nlm.nih.gov
>>
Caution with the Use of Recombinant
Activated Factor VII in Treating Postoperative
Hemorrhage in Cardiac Surgery
To the Editor:
We congratulate Bishop and colleagues for investigating the use of recombinant factor VIIa (rFVIIa) as a “rescue therapy for uncontrollable postoperative haemorrhage after cardiac surgery,” which is a challenging and potentially life-threatening complication. However we would like to raise the following 3 issues:
1. Since the Food and Drug Administration (FDA) first licensed rFVIIa, in 1999 for use in patients with hemophilia, its use has been expanded to include bleeding in non-hemophilia patients, but without a clear consensus on the clinical indications. Furthermore, concerns have been raised about its safety and efficacy in the latter group of patients with several reported cases of thromboembolic adverse events in the literature. These include cerebrovascular accidents, myocardial infarction, peripheral arterial thrombosis, pulmonary embolism, and deep venous thrombosis.
There is also evidence in the literature that such adverse events due to “off-label” use of rFVIIa are underreported. A recently published systematic review of literature about pharmacologic uses of rFVIIa identified 28 clinical trials in addition to about 300 case series and reports. It demonstrated that the use of rFVIIa was safe and effective in more than 90% of patients with hemophilia and other coagulation disorders. However the authors stated that there was not enough evidence to support the use of rFVIIa in patients without pre-existing coagulation disorder presenting with severe bleeding with or without surgery.
2. The cause and effect relationship is not clear in the study. The pre-rFVIIa coagulation screen was performed between the first cycle and the second cycle of non-red cell blood product support according to the management protocol of the cases reported. However, the post-rFVIIa coagulation screen was performed after both the second cycle of non-red cell blood product support and rFVIIa had been given. It is our contention that before any conclusions can be made about the cause and effect relationship between rFVIIa and the control of postoperative bleeding in this study, there should have been a control group consisting of patients who had been administered a second cycle of non-red cell blood product support, but without rFVIIa for comparison.
3. A particularly important safety concern with the use of rFVIIa in cardiac surgery is the potential for inappropriate clotting. After cardiac surgery with cardiopulmonary bypass, there is upregulation of tissue factor expression both locally in areas of tissue injury as well as systemically. Given that the mechanism of action for rFVIIa involves binding to tissue factor, therefore increased tissue factor expression may lead to more systemic clot formation. Furthermore, many cardiac surgery patients have vulnerable atherosclerotic plaques in their coronary vasculature; therefore excessive thrombin generation by rFVIIa may increase the incidence of acute coronary syndromes during the perioperative period.
We believe that a degree of caution should be applied regarding the use of rFVIIa in the control of excessive surgical bleeding after cardiac surgery, particularly in absence of both clear guidelines and evidence from clinical randomized trials.
Sharif Al-Ruzzeh, PhD, FRCS
Amr Mahmoud, MS, FCARCSI
Samir Shah, MS, FRCS
David O’Regan, MD, FRCS
The Yorkshire Heart Centre
Leeds General Infirmary
Leeds, LS1 3EX United Kingdom
REPLY
To the Editor:
My co-authors and I are greatly appreciative of the comments by Al-Ruzzeh and colleagues regarding the use of recombinant activated factor VII (rFVIIa) in the treatment of postoperative hemorrhage in cardiac surgery. As they point out, factor VIIa was used as rescue therapy for uncontrollable postoperative hemorrhage, which is indeed challenging and life threatening. The points they make are appropriate and valid.
1. The use of rFVIIa in our study was entirely off-label. We are indeed concerned about the possibility of intravascular thrombosis, and we were aware of reports in the literature of cerebrovascular accidents, myocardial infarction, and peripheral thrombosis. However, the overall reported rate of major adverse events was approximately 1%, although it is certainly probable that these complications are underreported. However, the reality in the clinical setting we faced was that the patients were bleeding profusely from a generalized coagulopathy, and the practical alternatives were more blood products, the use of rFVIIa, or eventual exsanguinations. The patients treated had rFVIIa and had predominantly complex ascending aorta and arch aneurysms (7 of 12), 5 of which were acute dissections, all with complex repairs partly performed under hypothermia and circulatory arrest. Each of these 12 patients had massive blood product replacement (mean of 19 units of fresh-frozen plasma, 23 units of platelets, 20 units of cryoprecipitate) prior to the decision to use rFVIIa. The alternative of using even more blood products was limited by blood bank resources, and uncertainty about outcome with the potential for ongoing hemorrhage.
2. Al-Ruzzeh and colleagues correctly point out that the pre-rFVIIa coagulation screen in the management protocol for excessive bleeding was performed between the first and second cycles of non-red cell blood product support. They suggest that following the second cycle of blood products, there should have been a control group of patients who did not receive rFVIIa versus those who did, so that a valid comparison could be made, thus enabling a better conclusion regarding cause and effect. This is logical from a theoretical point of view, but the reality was that these patients were bleeding despite massive coagulation factor replacement (on the average, double that recommended in our management protocol) and a further hemostatic intervention was desperately needed.
3. There is no doubt that rFVIIa is incredibly effective in promoting coagulation. We did not observe any thrombotic-related complications. This may have been due to the fact that the coagulopathy was extreme in the first place, but this is purely conjecture. One would certainly be concerned about the fate of any coronary bypass grafts. Major thoracic aorta and arch aneurysm (resection and repair), particularly in the setting of acute dissection continues to have a high perioperative mortality, at least in part related to uncontrolled bleeding from suture lines, despite the use of good hemostatic techniques, including felt and bovine pericardium buttresses, Bioglue, and aprotinin. It is frustrating and disappointing to see ongoing nonsurgical hemorrhage after a long complex and anatomically successful reconstruction, particularly after the use of massive coagulation product replacement.
If the alternative is eventual exsanguination and death, then it would seem reasonable to point out that the use of rFVIIa can result in impressive hemostasis, and that the cardiac surgeon can offer the patient potential salvage in these dire circumstances. We were pleased that all 12 patients in the report were discharged alive from the hospital and were alive at follow-up 30 to 90 days later. Nevertheless, we certainly do acknowledge and recommend that a significant degree of caution should be applied regarding the use of rFVIIa, even in this desperate setting.
James Tatoulis, MB, BS
The Royal Melbourne Hospital
University of Melbourne
Melbourne, Victoria, 3050 Australia
<< |
