The use of recombinant factor VIIa in the treatment of bleeding disorders

Prothrombin complex concentrates (PCCs), containing prothrombin and factors VII (fVII), IX, and X, were initially developed for treatment of hemophilia B (factor IX deficiency). These concentrates were known to contain small amounts of activated factors, particularly factors VIIa, IXa, and Xa. Their presence led to the use of PCCs in hemophilic patients with inhibitors to factors VIII or IX with the idea that the activated factors might provide a way to “bypass” the need for factor VIII or IX and represent a new approach to the treatment of inhibitors.¹ 

A clinical trial of the effectiveness of PCCs compared with an albumin placebo was carried out in 1980 and showed that PCCs were significantly more effective in controlling bleeding in hemophilic patients with inhibitors than an albumin placebo.²  On the other hand, replacement therapy with factor VIII for similar bleeding episodes in noninhibitor hemophilia patients was virtually 100% effective. Thus, very early on it was known that the term “inhibitor bypassing agent” actually described partial rather than complete effectiveness.

The effectiveness of PCCs in hemophiliacs with inhibitors raised the question as to which component(s) of PCCs was effective as a hemostatic agent. As early as 1975, it was noted that concentrates with higher amounts of factor VII seemed to be more effective than those with lower concentrations of this factor. However, the idea that factor VII or factor VIIa alone might be an active component in PCCs was not pursued at that time.

Since activated factors in PCCs were thought to be responsible for the hemostatic effect, “activated PCCs” were prepared commercially. Autoplex (Baxter Bioscience, Glendale, CA) and FEIBA (factor 8 inhibitor bypassing agent) (initially prepared by Immuno, Vienna, Austria, but now by Baxter Bioscience) were prepared by so-called “controlled activation” of the original “unactivated” prothrombin complex concentrates.³  In clinical trials, the activated PCCs were equal to or marginally better than the original PCC preparations, though the active ingredients remained unknown.^3,4 

Factor VIIa was considered by several investigators to be a candidate for the major bypassing agent.⁵  In 1981, Kingdon and Hassel injected Autoplex, which contained several activated factors, into a hemophilic dog in which bleeding was induced by a standardized surgical incision on the gum.⁶  This treatment was found to be partially effective in establishing hemostasis in these animals. Although several investigators suspected that factor VIIa possessed bypassing activity, it remained for Hedner and Kisiel to clearly establish the effectiveness of factor VIIa as a hemostatic agent in humans.⁷  They purified factor VII from human plasma under sterile conditions. During preparation, factor VII was activated to factor VIIa. The first preparation was administered to 2 hemophilic patients who had high-titer-factor factor VIII antibodies and who were experiencing bleeding episodes.⁷  Bleeding in both patients was controlled by the administration of plasma-derived factor VIIa. Subsequently, plasma-derived factor VIIa was observed to be effective in controlling hemorrhage in other hemophilic patients with inhibitors.⁸  The difficulty in preparing factor VII from plasma resulted in a shortage of starting material that precluded extensive use of plasma-derived factor VIIa. The problem of availability was solved by the development of recombinant factor VIIa (rfVIIa) by Novo Nordisk Pharmaceuticals (Bagsvaerd, Denmark).

Prior to use in human subjects, recombinant factor VIIa was administered to hemophilia A and B dogs subjected to a standardized toenail bleeding time, which was used as an in vivo measure of thrombin generation. Dogs with severe hemophilia A were given the recombinant product in doses ranging from 50 to 220 μg per kg/body weight (BW).⁹  The bleeding time in these dogs was normalized except in one animal, which was given the lowest dose of the factor. Similar success was observed in dogs with hemophilia B.⁹  Hemostasis in dogs homozygous for von Willebrand disease (VWD) was not achieved even at higher doses of factor VIIa, suggesting early on that platelet adhesion and aggregation were necessary for the full action of rfVIIa.⁹  Unlike humans, canine platelets do not contain von Willebrand factor (VWF).

Pharmacokinetic studies in adults revealed a circulating half-life (T1/2) of rfVIIa to be 2.60 to 2.84 hours that was independent of dose.¹⁰  The recovery was slightly less in bleeding versus nonbleeding patients. In the pediatric age group the T1/2 was 1.32 hours, significantly shorter than that seen in adults. The clearance rate was also more rapid in children at 67 mL per kg/h, suggesting that higher doses of factor VIIa might be required in the pediatric age group.¹¹ 

The mechanism of action of fVIIa is controversial, but there is general agreement on several points. It is generally agreed that about 1% of circulating factor VII in healthy individuals is in the activated form and that the amount of the factor VIIa required for “bypassing activity” is much larger than this.^12-15  It is also generally agreed that the tissue factor (TF) pathway must be intact for fVIIa to have an effect.^14,16,17  The major disagreement appears to revolve around the issue of whether rfVIIa has any effect independent of tissue factor.¹⁷  One group has developed a cell-based model system of coagulation to study the mechanism of action of fVIIa. In this system, all the plasma procoagulants and relevant inhibitors are added to tissue factor-bearing cells and unactivated platelets in microtiter wells in which coagulation is initiated by addition of fVIIa and calcium.¹⁸  Factor VII and fVIIa are equally effective at initiating the process.¹⁹  In this system, platelet activation occurs first, followed by a burst of thrombin generation. The model system can also be used to mimic the hemophilic condition by omission of factor VIII or IX. A schematic representation of the normal and hemophilic conditions are shown in Figure 1A-B. In the absence of factor VIII or IX it can be shown that at pharmacologic doses, fVIIa will bind weakly to activated platelets and can directly activate factor X to factor Xa such that in the presence of factor V, thrombin generation can be improved over and above that generated by the tissue factor/VIIa complex alone (Figure 2).^20-22  The binding of rfVIIa to activated platelets may explain why rfVIIa is localized only to the site of bleeding. In this model, it is clear that the tissue factor pathway must be intact in order for the initial activation of platelets to occur by the small amount of thrombin formed at the level of the tissue factor-bearing cell. In addition, the cell-based model system predicts that there is a dose response relationship between fVIIa and thrombin generation. As can be seen in Figure 2, increasing doses of rfVIIa can increase thrombin generation in this system, even in the absence of either factor VIII or factor IX. However, thrombin generation in this system is not completely normalized even at higher rfVIIa concentrations.

By contrast, Mann and colleagues^23-26  hold the view that the mechanism of action of factor VIIa is dominated by the tissue factor-mediated activity with only a minimal, if any, contribution from TF-independent fVIIa activity. In studies using TF incorporated into lipid vesicles, van't Veer et al observed that fVIIa overcomes the inhibitory effect of zymogen fVII on thrombin generation.^23,24  While the influence of rfVIIa is dependent on TF, Butenas et al observed that phospholipids or platelets substantially increase the hemostatic potential of rfVIIa.^25,26 

Lisman and De Groot have recently analyzed the alternative mechanisms of action of rfVIIa.¹⁵  They conclude that the tissue factor pathway is required for rfVIIa to act and that at pharmacologic doses, rfVIIa will also bind to activated platelets and directly activate fX to fXa as suggested by the Roberts group.^12-14  Lisman et al have also observed that the enhanced thrombin generation from fVIIa not only accelerates clot formation, but also inhibits fibrinolysis by activation of thrombin activatable fibrinolytic inhibitor (TAFI) in factor VIII-deficient plasma.²⁷ 

Whatever the mechanism of action, it appears that the action of rfVIIa is localized to the site of bleeding since systemic coagulation, if it occurs, appears to be minimal despite the administration of large doses of the factor, even in people without coagulopathy.

Due to the lack of a specific laboratory test to monitor the effect, initial doses of rfVIIa were chosen on the basis of subjective evaluation of the patient's response. In a randomized double-blind trial of 35 and 70 μg per kg/BW, carried out in 84 subjects with hemophilia A and B with and without inhibitors, it was found that both doses of rfVIIa were about 70% effective in patients with hemarthroses and soft tissue bleeding.²⁸  The mean number of doses required to achieve a subjective response was about 3 for both dose regimens with doses given every 2 to 3 hours. However, in a later prospective randomized trial in hemophilia patients with inhibitors undergoing elective surgery, 35 and 90 μg recombinant factor VIIa were administered. This study showed that the higher dose was more effective than the lower dose.²⁹  More recently, however, doses of 120 μg per kg/BW have been used routinely.³⁰  Given the short T1/2 of factor VIIa, doses have been administered intravenously every 2 hours for the first 48 hours and every 2 to 6 hours for the next 72 hours in nonsurgical cases. Achievement of hemostasis has varied, but in later studies up to 90% of patients have responded to 90 to 120 μg/kg doses during the first 48 hours.³¹  Even though the recommended dose is 90 μg/kg, it is clear that the optimal dose and dosing intervals of factor VIIa have not been established with certainty. Recent developments suggest that doses higher than those recommended may be more efficacious in patients with hemophilia and may require dosing at less frequent intervals.^32,33 

Constant infusion regimens for the use of rfVIIa have also been described and have been used successfully.³⁴  With this regimen it may be possible to use less rfVIIa when compared with treatment regimens using bolus doses. However, the continuous infusion regimen is currently not in wide use. When rfVIIa is given as a constant infusion, the peak of thrombin generation is thought to be lower than that seen with bolus doses of rfVIIa.³⁵ 

Using the cell-based model of coagulation to mimic in vivo coagulation conditions, it was shown that increasing doses of factor VIIa resulted in increased thrombin generation as shown in Figure 2. In an attempt to increase thrombin generation, these studies led to the use of higher than the recommended doses of rfVIIa for hemophilic patients. In one of the earlier case reports, a 14-year-old patient with hemophilia B who developed a target joint in the left elbow did not respond to the usual doses of factor VIIa but did respond to a dose of 320 μg per kg/BW.³²  This dose was given with the knowledge that the T1/2 of factor VIIa in the patient was 1.5 hours, with a clearance rate nearly double that of the average adult.

The effectiveness of recombinant factor VIIa at doses of 300 μg/kg has been tested in 3 young patients with hemophilia A with inhibitors to factor VIII.³³  Of 114 bleeding episodes, 95 responded to a single 300-μg dose of rfVIIa without side effects. The remaining patients responded to subsequent similar doses of rfVIIa. The bolus injection of this mega dose was considered superior to a standard dose or to a constant infusion of regular doses of the product. Although the number of patients in these studies is small, the data suggest that higher doses are not only more effective but are also safe.³³  It has been noted that with higher doses of VIIa, peak thrombin generation is higher and is thought to result in a more stable and less porous fibrin clot.^36,37 

It has also been reported that sequential therapy with factor VIIa and prothrombin complex concentrates may be more beneficial than factor VIIa alone, suggesting a synergistic effect between the 2 agents.³⁸ 

The “mega” doses of rfVIIa used in these studies have not been approved by the FDA and represent “off-label” use of the product. Furthermore, these high doses of rfVIIa have not been used extensively in elderly patients and in those who do not have hemophilia.³⁹ 

Recombinant factor VIIa has been approved by the FDA in the United States only for use in hemophilia A and B patients who have inhibitory antibodies against factor VIII or factor IX. In Europe, rfVIIa has been approved not only for hemophilia with inhibitors, but also for postsurgical bleeding, hemophilia home treatment, factor VII deficiency, and Glanzmann thrombasthenia. The recommended dose according to the package insert is 90 μg per kg/BW of recombinant factor VIIa given in bolus doses intravenously. It is recommended that doses be given every 2 hours until bleeding ceases. Response rates have generally been excellent, ranging up to 90% effectiveness in bleeding episodes in patients with inhibitors. The importance of early therapy has been stressed.⁴⁰  As discussed in the previous section, dosage requirements differ depending on the circumstances surrounding a given bleeding episode. Factor VIIa has also proved effective for the home treatment of bleeding episodes in hemophiliacs with inhibitors. In a “home therapy” trial encompassing 56 patients and 827 bleeding episodes, it was found that the product was effective in 90% of cases.⁴¹ 

Recombinant factor VIIa has also been used to achieve hemostasis for major surgery in hemophilic patients with inhibitors. Again, dose regimens have varied, but using bolus doses, Ingerslev has recommended that patients undergoing major surgery receive 90 to 120 μg per kg/BW every 2 to 3 hours of factor VIIa throughout the first 2 to 3 days following surgery.⁴²  On the third and fourth days, dosage intervals may be decreased to every 3 to 4 hours. Thereafter dosage intervals can be decreased to every 4 to 6 hours until about 7 days. If bleeding complications occur after initial hemostasis is achieved, factor VIIa may be given again every 2 hours until bleeding ceases. In most cases, antifibrinolytic agents are given concomitantly with rfVIIa.⁴² 

Failure to respond to rfVIIa may be due to suboptimal dosing. The major problems in using rfVIIa in hemophilia patients are related to dosage, requirement for frequent dosing, and the lack of a quantitative method for monitoring therapeutic effectiveness.

Factor VIIa has also been used successfully to treat patients with factor VII deficiency. Bleeding episodes in these patients have responded to relatively low doses of the product, ranging from 15 to 20 μg per kg/BW every 2 to 3 hours until cessation of bleeding.^43,44  Major surgery has also been carried out in factor VII-deficient patients under cover of rfVIIa. Only one factor VII-deficient patient has developed anti-factor VII antibodies after treatment with rfVIIa.⁴⁵ 

Recombinant VIIa has also been used to treat patients with factor XI deficiency either with or without inhibitors to factor XI.⁴⁶  Doses of 90 to 120 μg/kg every 2 to 3 hours until bleeding ceases have been found to be effective in this condition, but the optimal dose has not been clearly defined. Some investigators now consider rfVIIa to be the treatment of choice in factor XI deficiency and for inhibitors to factor XI.

There are anecdotal reports of the successful use of the product in stopping bleeding in humans with von Willebrand disease, including type III.⁴⁷  Factor VIIa has been used in those patients who do not respond to conventional therapy because of antibodies to von Willebrand factor (VWF) or due to other complications such as angiodysplasia. Bleeding resulting from oral surgery and angiodysplasia in type III VWD, complicated by an alloantibody against VWF, reportedly responded to bolus infusion of rfVIIa.⁴⁸  Some of these patients also received antifibrinolytic agents as well as von Willebrand factor products.

There are anecdotal reports of the use of factor VIIa for the treatment of bleeding episodes in patients with Glanzmann thrombasthenia, mostly in those patients who have become unresponsive to platelet transfusions because of the development of alloantibodies to glycoproteins (GPIIb/IIIa). Many of these cases have been reported by Poon et al, who have established a registry of the use of factor VIIa in patients with congenital platelet disorders.⁴⁹  Recombinant VIIa given in doses ranging from about 70 to 100 μg/kg every 2 hours until bleeding stops has been effective in controlling bleeding in many cases of Glanzmann thrombasthenia. In one series of patients, the use of factor VIIa was reported to be effective in about 48% of patients.⁵⁰ 

Factor VIIa has also been used successfully in Bernard-Soulier syndrome by a few patients unresponsive to platelet transfusion or desmopressin (DDAVP).⁵¹  Factor VIIa has also been reported to be successful in the treatment of hemorrhages in a patient with the Hermansky-Pudlak syndrome.⁵²  More experience is necessary before the efficacy of fVIIa can be clearly established in patients with congenital thrombocytopathy.

In anecdotal reports, bleeding in patients with thrombocytopenia who have not responded to conventional therapy have responded to rfVIIa in doses of 50 to 100 μg per kg/BW. There are reports that the bleeding time was reduced in more than 50% of the patients, and in the majority of cases, bleeding was controlled.^53,54  For factor VIIa to be effective in patients with thrombocytopenia, the minimal platelet count should be in the range of 5000 to 30 000/μL. More data are needed before one can adequately assess the efficiency of rFVIIa in thrombocytopenic disorders, although initial reports appear promising.

Recombinant factor VIIa has been used for routine bleeding that has failed conventional therapy in a large number of medical disorders. Many of the reports are anecdotal and represent off-label use of the product. Several trials of rfVIIa are ongoing, including prophylaxis for hemophilia with inhibitors, treatment of bleeding after spinal surgery, and orthotopic liver transplantation. There are completed clinical trials for use of rfVIIa in intracranial hemorrhage and in bleeding associated with trauma, but official results from these trials are not yet available. A trial for use of rfVIIa to control bleeding in cardiopulmonary bypass is planned. In the following paragraphs, some of the more common uses of rfVIIa in bleeding problems secondary to medical conditions will be discussed.

Factor VIIa has been used for bleeding in patients with liver disease, including patients who have experienced variceal bleeding from the esophagus and stomach.^55,56  Doses have varied, ranging from 5 to 120 μg per kg/BW.⁵⁷  The response to infusion of rfVIIa has varied, but in one single-center study, 10 patients with alcoholic cirrhosis and esophageal bleeding were given 80 μg rfVIIa and observed over a 12-hour period. Bleeding was controlled immediately after treatment and lasted for 12 hours in 5 patients. Factor VIIa has been used prophylactically in patients undergoing laparoscopic liver biopsy.⁵⁸  Reports indicate that at least 74% of patients treated with factor VIIa have maintained hemostasis.

In one controlled trial, a total of 245 patients with cirrhosis and gastrointestinal hemorrhage were treated with 8 doses of rfVIIa at 100 μg per kg/BW or with placebo. The end-point was control of gastrointestinal bleeding within 24 hours after rfVIIa administration. In patients in the Childs-Pugh B and C subgroups, those receiving rfVIIa were most likely to exhibit control of bleeding than patients on placebo. However, patients in the Childs-Pugh A subgroup receiving rfVIIa were similar to those on placebo.⁵⁵ 

There have been other anecdotal reports of the successful use of rfVIIa in the control of bleeding in patients with liver disease, but failures as well as recurrent bleeding have been reported. In most patients, rfVIIa has been more effective than plasma therapy alone but plasma replacement was synergistic with rfVIIa.⁵⁷  Factor VIIa has also been used successfully to control bleeding during liver transplantation and during partial hepatectomy in noncirrhotic patients.⁵⁹ 

Factor VIIa has also been used successfully in patients with intracranial hemorrhage as first reported in hemophilia patients with inhibitors.⁶⁰  However, recent experience suggests that intracranial hemorrhage in patients with normal coagulation may also benefit from early therapy with fVIIa.⁶¹  There is evidence that after intracerebral hemorrhage occurs there is expansion of the hematoma mass in more than 50% of patients for up to 6 hours. These observations suggest that early arrest of hematoma growth would improve outcome, and rFVIIa has been suggested as an agent for early intervention.

The administration of rfVIIa to patients with warfarin intoxication results in marked reduction of the prothrombin time in doses ranging from 15 to 90 μg/kg, but perhaps more importantly, rFVIIa enhanced hemostasis in those patients who were experiencing hemorrhage.⁶²  Recombinant factor VIIa shortens the prothrombin time (PT), but this is not correlated with cessation of hemorrhage.

It has been suggested that rfVIIa is effective in reversing the effect of low-molecular-weight heparin (LMWH).⁶³  On the other hand, it has shown that rfVIIa does not reverse the anticoagulant effects of low-molecular-weight heparin (LMWH) in rabbits, but the dose of LMWH in this study may have been excessive.⁶⁴  Since it has been shown that rfVIIa will not correct the PT or partial thromboplastin time (PTT) of plasma completely deficient in factor X, it may not reverse the effect of factor Xa inhibitors such as low-molecular-weight heparin if factor Xa was completely inhibited.⁶⁵  It is not known whether rfVIIa will be useful in reversing the effects of argatroban and hirudin.⁶⁶  Factor VIIa has been shown to be effective in reversing the anticoagulant effect of the nematode anticoagulant peptide.⁶⁷ 

Factor VIIa has also been used in many other medical conditions such as bleeding from renal failure,⁶⁸  factor X deficiency and amyloidosis,⁶⁹  diffuse alveolar hemorrhage,⁷⁰  and bleeding following extensive burns.⁷¹  This list is not exhaustive, and there are hundreds of anecdotal reports on the efficacy of VIIa in controlling hemorrhage in a variety of medical disorders when other measures have failed.⁷¹  However, the product is not always effective in controlling bleeding whether due to inadequate dosing, simple failure of the agent, or to unknown other causes.

One of the first off-label uses of rfVIIa was in a soldier who suffered extensive bleeding following a gunshot wound to the abdomen.⁷²  This patient had extensive abdominal surgery to control bleeding and required numerous transfusions of red cells, plasma, cryoprecipitate, and platelets. As a result, he experienced hemodilution, hypothermia, and a severe coagulopathy characterized by thrombocytopenia, prolonged PT and PTT, and hypofibrinogenemia. In an attempt to control hemorrhage, the patient was given 60 μg per kg/BW of rfVIIa on 2 occasions with subsequent dramatic control of the hemorrhage.⁷² 

This amazing result led to a series of anecdotal reports on the use of rfVIIa in uncontrolled bleeding conditions in patients subjected to extensive trauma.⁷³  In one study in patients with massive trauma and a multifactorial coagulopathy, 19 patients were treated with 1 to 3 doses of rfVIIa, and in 15 patients excessive hemorrhage was controlled, sometimes within minutes after administration of rfVIIa.⁷⁴  However, 4 of the 19 patients did not respond to rfVIIa.⁷⁴  Similar results have been observed in other critically ill trauma patients.^75,76 

Patients with severe acidosis or profound shock seem less likely to respond.⁷⁷  A randomized, placebo-controlled trial of the use of rfVIIa in trauma patients has recently been completed but the results are not yet available.

Recombinant fVIIa has also been effective in controlling excessive hemorrhage in patients undergoing cardiac surgery.⁷⁸  In anecdotal reports, administration of 1 to 2 doses of rfVIIa at 90 to 120 μg per kg/BW has resulted in effective hemostasis, again without evidence of systemic activation of the clotting system.⁷⁹  Despite reports of the successful use of rfVIIa, there are reports suggesting that it is not always effective.⁸⁰ 

FVIIa has been used to arrest bleeding in children with large and disfiguring facial hemangiomas.⁸¹  When used prophylactically in these conditions, blood loss is reduced and the surgical field is relatively bloodless. None of the children has suffered any serious adverse events from administration of rfVIIa.

In a double-blind trial of rfVIIa of healthy patients undergoing prostatectomy, it was found that the administration of 40 μg/kg rfVIIa reduced blood loss to the extent that the need for transfusion of red cells was eliminated.⁸² 

The need for controlled clinical trials that address some of the questions related to indications for rfVIIa, type of surgery, dose, and potential side effects is also recognized. For the most part, except for 2 controlled trials related to the use of rfVIIa in patients undergoing prostatectomy and partial liver resection, reports on the efficacy of rFVIIa for surgical bleeding are anecdotal and represent off-label use of the product.

There is continuing concern that factor VIIa may cause thromboembolic side effects largely because it is an activated factor given in doses to raise fVIIa levels more than 1000-fold. Evidence to date, however, suggests that rfVIIa is both safe and effective both in hemophilic and nonhemophilic conditions.

It is likely that the safety of the product is due to its selective location at the site of bleeding. As of early 2004, more than 700 000 doses of factor VIIa have been administered to several thousand hemophilic patients with inhibitors with about a 1% incidence of serious adverse events, including myocardial infarction, stroke, pulmonary embolism, and deep venous thrombosis among others.³⁵  Similar adverse events have been reported in a small number of nonhemophilic patients, but the true incidence is not known. Many of the thrombotic events have occurred in patients with a predisposition to thrombotic complications such as diabetes mellitus, obesity, cancer, and atherosclerotic cardiovascular disease, and administration of rfVIIa to such patients should be approached with caution. Adverse events have not been related to dose. Although FVIIa has an excellent safety profile, further monitoring of safety both in the hemophilic and nonhemophilic condition is warranted.

There currently is no satisfactory laboratory test to monitor the clinical effectiveness of rfVIIa. With hemophilic and nonhemophilic patients, administration of FVIIa results in shortening of prothrombin time and the partial thromboplastin time. The prothrombin time generally shortens to around 7 to 8 seconds except in factor V- or factor X-deficient plasma, suggesting that patients completely deficient in factors V and/or X will not benefit from therapy with this product.⁶⁵  The partial thromboplastin time shortening has been demonstrated to be due to the direct activation of factor X by circulating factor VIIa on the phospholipids used in the partial thromboplastin time test.⁶⁵  Thus, shortening of these 2 screening tests of coagulation does not necessarily reflect clinical effectiveness, which is judged subjectively.

Methods being evaluated for monitoring include thromboelastography, thrombin generation over time, wave form analysis of the PTT, and platelet contractile force.⁸³ 

There is general agreement that factor VIIa is safe and effective in hemophilia patients with inhibitors. Emerging data from anecdotal reports and completed or ongoing clinical trials suggest that rfVIIa is also effective in a variety of bleeding conditions in nonhemophilic patients. This has led some to use the term “universal” or “general” hemostatic agent. Perhaps the use of this term is unfortunate since rfVIIa is not always effective in all hemorrhagic conditions. However, rfVIIa has been shown to be life saving in some patients experiencing life-threatening hemorrhage. In such cases, the terms universal or general become irrelevant, and use of factor VIIa in such patients seems warranted even in the absence of controlled clinical trials.