Recombinant factor VIIa: Administration and adverse effects

INTRODUCTION — Unlike most other clotting factor products, which are inactive precursor proteins, recombinant activated factor VII (rFVIIa) is an activated form of clotting factor. Thus, it carries greater risks of thrombosis than other factor products in some settings, and clinicians must incorporate clinical judgment to balance the potential benefits of administration with this increased risk.

This topic review discusses clinical use of rFVIIa, including potential benefits, dosing, and adverse effects, with a focus on off-label indications.

Use of rFVIIa in the treatment or prevention of bleeding in individuals with hemophilia A or B with an inhibitor is discussed separately. (See "Treatment of bleeding and perioperative management in hemophilia A and B", section on 'Inhibitors'.)

BACKGROUND AND MECHANISM OF ACTION

Background and rationale for development — The rFVIIa product was originally developed for people with hemophilia who could no longer use factor replacement products because they had developed an inhibitor (antibody against infused factor).

Management of patients with hemophilia A or B can be complicated by the development of inhibitors. It is not possible to neutralize high titer inhibitors by administration of very high levels of replacement therapy. For this reason, it was a high priority in the 1970s and 80s to develop products that could "bypass" the need for the deficient factor and establish hemostasis in people with hemophilia with inhibitors. Such products are referred to as "bypassing agents."

●PCC – The first product identified was prothrombin complex concentrate (PCC), a complex of vitamin K-dependent clotting proteins (factors II, VII, IX, and X, and protein C and S) that co-purified from normal plasma by methods available at the time. PCC was initially used as a source of factor IX to treat individuals with hemophilia B. It was subsequently noted to improve hemostasis in a patient with hemophilia A who had developed an inhibitor [1].

●aPCC – Subsequently, products containing increased levels of activated coagulation factors (activated prothrombin complex concentrates [aPCCs]) were prepared in an attempt to increase the efficacy of bypassing therapy for individuals with hemophilia A with an inhibitor [2-4]. The only product of this type remaining on the market is factor eight inhibitor bypassing agent (FEIBA). The aPCCs were indeed more effective than unactivated PCC, but they were associated with an increased risk of thrombosis [5].

Several components in these products were thought to contribute to their thrombogenicity: non-vitamin K-dependent zymogens, activated factors IX and X, and phospholipids from platelets resulting from inadequate centrifugation of the donor plasma. On the other hand, factor VIIa was hypothesized to play a major role in the hemostatic efficacy of aPCC in people with hemophilia and inhibitors.

●Plasma-derived factor VIIa – Purified, plasma-derived factor VIIa was tested in the late 1980s in an attempt to develop a less thrombogenic bypassing agent [6,7]. When used at doses of 5 to 20 mcg/kg, it was found to be hemostatically effective for surgery and joint bleeding, without thrombotic complications or evidence of systemic activation of coagulation.

●rFVIIa – A recombinant factor VIIa (rFVIIa) product was developed and likewise shown to be effective as a bypassing agent in people with hemophilia with inhibitors and in acquired hemophilia (development of an autoantibody to endogenous factor VIII) [8-13]. It appeared to require doses >90 mcg/kg for efficacy in most patients, with no evidence of systemic activation of coagulation. However, subsequent studies have shown that rFVIIa can be associated with an increased risk of thrombosis in some settings, particularly in patients who do not have hemophilia. (See 'Thromboembolic complications' below.)

The first rFVIIa product was generated from cultured cells, and the second from rabbit mammary gland secretion. (See 'Available products and their properties' below.)

Overview of rFVIIa mechanism — High-dose rFVIIa was originally thought to act by increasing the activity of the extrinsic, tissue factor (TF)-associated, coagulation pathway, as occurs with physiologic levels of factor VIIa at sites of injury [14].

However, when administered at pharmacologic doses, rFVIIa binds to the surface of activated platelets in a TF-independent manner and promotes factor X (FX) activation and thrombin generation on the activated platelet surface [15-17]. Binding of factor VIIa to platelets appears to involve the glycoprotein Ib/IX/V complex and anionic phospholipids expressed on activated platelets [18,19]. Additional mechanisms may contribute to platelet binding and rFVIIa mechanism of action [20,21].

In the hemophilias, platelet-bound rFVIIa partially restores platelet-surface FX activation, which is deficient because of the absence of factor VIIIa/IXa complexes. In nonhemophilic conditions, platelet-bound rFVIIa increases activation of both factor IX and FX and increases thrombin generation above normal levels [22]. Increased thrombin generation then promotes increased activation and local accumulation of platelets (functional or dysfunctional platelets), potentially improving hemostasis in a wide range of bleeding conditions [8].

Clinical observations and data from experimental models support this platelet-dependent and TF-independent mechanism for the hemostatic effect of rFVIIa. For example:

●The high concentrations of rFVIIa required for hemostatic efficacy in persons with hemophilia are far greater than would be required to saturate TF [14].

●In vitro, factor VIIa binds to activated platelets in a TF-independent manner and enhances platelet-surface thrombin generation in the presence of plasma levels of coagulation factors [23].

●Platelets from healthy volunteers and individuals with Glanzmann thrombasthenia can support fibrin generation via rFVIIa-mediated thrombin generation independent of TF [24].

●In a mouse model of hemophilia B (factor IX deficiency), a form of factor VIIa that is unable to bind to TF can control bleeding as effectively as normal mouse factor VIIa [25].

A potential role for TF-dependent activity in the hemostatic effect of rFVIIa cannot be completely ruled out [26-29]. This is especially true in certain off-label uses. Because there is a very large amount of TF in the brain, rFVIIa may indeed act by a TF-dependent mechanism when used to control intracerebral hemorrhage [30].

rFVIIa may also have TF-independent effects on endothelial cells. Factor VIIa binds to endothelial protein C receptor (EPCR) [31]; this interaction contributes to the hemostatic activity of rFVIIa in mouse models [21,32,33]. This effect may be due to release of endothelial extracellular vesicles that can support thrombin generation [21]. rFVIIa binding to EPCR can also trigger antiinflammatory responses and enhance endothelial barrier function [34]. It is not known whether these activities contribute to the effects of rFVIIa in humans.

Implications for clinical use — The mechanism of action of rFVIIa has important implications for its clinical use(s):

●The doses that are used clinically would not be expected to saturate platelet binding. This means that increasing the dose (increasing the plasma concentration) of rFVIIa should increase the rate of platelet-surface thrombin generation. Therefore, some practitioners have used doses of rFVIIa well above the 90 mcg/kg dose recommended by the manufacturer, and they feel that this has resulted in increased hemostatic efficacy in some patients with hemophilia.

●Localization of rFVIIa activity to the activated platelet surface appears to explain the low rate of thrombotic complications, particularly in patients with hemophilia. Factor VIIa-driven thrombin generation occurs on the appropriate surface (activated platelets) and is subject to most of the normal control mechanisms that prevent inappropriate coagulation [15].

●There is significant inter-individual variability in the hemostatic response to factor VIIa due to levels of other coagulation factors (eg, fibrinogen, prothrombin, factor X) as well as platelet number and function [35-38]. Therefore, the appropriate dose and effectiveness of rFVIIa may differ depending on a number of circumstances [39].

●Changes in body temperature and pH may reduce the activity of factor VIIa. As an example, in one study, rFVIIa activity decreased by approximately 90 percent and tissue factor (TF)-rFVIIa complex activity decreased by approximately 60 percent when pH changed from 7.4 to 7.0 [40]. Thus, acidosis may markedly impair hemostatic function and reduce the response to rFVIIa.

●The half-life of rFVIIa in the circulation is two hours or less, shorter than that of normal factor VII (4 to 6 hours), as well as that of most of the other coagulation factors (table 1).

The above-listed circumstances may limit the effectiveness of rFVIIa, especially in surgical and trauma patients, who may have a combination of hypothermia, acidosis, electrolyte disturbances, and loss of platelets and coagulation factors due to consumption or dilution by fluid administration and/or massive transfusion [39].

AVAILABLE PRODUCTS AND THEIR PROPERTIES — In the United States, two rFVIIa products are available, NovoSeven and Sevenfact. Both are produced recombinantly, but they differ in their production and other important characteristics and cannot be considered interchangeable.

●NovoSeven – NovoSeven RT (NovoSeven, NiaStase RT) has the following characteristics [41]:

•Licensed in 1999

•Human factor VII sequence

•Generated in cultured hamster (BHK) cells grown in newborn calf serum (purified from culture supernatant)

•Converted to factor VIIa during purification

•Approved for several indications (no age restrictions):

-Hemophilia A or B with inhibitors

-Acquired hemophilia

-Glanzmann thrombasthenia

-Congenital factor VII deficiency

•Half-life 2.6 to 6 hours

●Sevenfact – Sevenfact has the following characteristics [42]:

•Licensed in 2020

•Human factor VII sequence

•Obtained from genetically modified transgenic rabbits (secreted in breast milk)

•Converted to factor VIIa during purification

•Approved for treatment of bleeding in hemophilia A or B with inhibitors (no age restrictions; was initially for individuals ≥12 years)

•Half-life 1.4 to 1.7 hours

Both products carry Boxed Warnings regarding the risk of thrombosis. (See 'Thromboembolic complications' below.)

An rFVIIa analog with enhanced activity (vatreptacog alfa) is no longer under investigation because of the development of anti-drug antibodies in a few patients [43].

GENERAL APPROACH TO ADMINISTRATION — Details of rFVIIa administration for specific bleeding disorders are discussed in the sections below. (See 'Approved indications' below and 'Off-label uses' below.)

The following considerations related to dose, interval, and duration of therapy apply generally and can be used to determine an individualized approach for indications not included below.

Overview of situations in which rFVIIa may be helpful — High-quality data on the efficacy and safety of rFVIIa are mostly available for approved indications. Data for off-label use are limited, but reports involving isolated patients or small series are numerous. Likewise, there is very limited experience to guide dose, dosing interval, or duration of therapy.

In most nonhemophilia applications, we generally consider rFVIIa to be appropriate only for patients who are experiencing life-threatening bleeding or are likely to experience life-threatening perioperative bleeding if not treated, and for whom hemorrhage has not responded to transfusion or other conventional therapies. Other criteria include an underlying bleeding condition for which no other therapy is available, lack of access to other therapies, or inability of the patient to tolerate other therapies. In short, we generally recommend that rFVIIa be administered when standard therapies appear to be failing or are unavailable.

It is difficult, if not impossible, to carry out a randomized clinical trial to study off-label use of rFVIIa when it is used in the manner we recommend. Such use is a relatively rare occurrence in most clinical settings and applies to a very heterogeneous group of patients. It would be quite difficult to obtain informed consent and accrue enough patients in this clinical situation to perform a reasonable analysis. Clinicians may not wish to have their patients randomized to receive no treatment or to continue receiving treatment that has been judged a failure.

Furthermore, many of these patients are so severely ill that they may have a bad outcome despite having received rFVIIa, even if it is an effective hemostatic agent. In short, it seems unlikely that it will be possible to acquire high-quality clinical data on the effectiveness of rFVIIa for most off-label uses. Therefore, we are only able to give general guidelines for such uses of rFVIIa.

Specific conditions for which rFVIIa has been reported to be effective include:

●Hemophilia A or B with an inhibitor

●Factor XI deficiency with an inhibitor [44]

●Acquired factor inhibitors (acquired hemophilia)

●Glanzmann thrombasthenia unresponsive to platelet transfusions (or to avoid platelet transfusions)

●Trauma or uncontrolled surgical bleeding [45-48]

●Bone marrow aplasia following hematopoietic cell transplantation or in aplastic anemia (along with platelet transfusions) [49,50]

●Liver disease [51]

●Certain surgical procedures (neurosurgery, cardiac surgery) [51]

●von Willebrand disease (VWD) with an inhibitory autoantibody against von Willebrand factor (VWF) [51]

Use has also been described in treating bleeding in patients receiving anticoagulants, but specific reversal agents are preferred when available. (See "Reversal of anticoagulation in intracranial hemorrhage" and "Management of warfarin-associated bleeding or supratherapeutic INR" and "Management of bleeding in patients receiving direct oral anticoagulants".)

Several meta-analyses have attempted to evaluate the efficacy in large populations with a mixture of underlying bleeding disorders (see 'Approved indications' below and 'General evidence for off-label use' below), with the caveats noted above. When balancing the risks and benefits of rFVIIa for an individual patient, it is important to realize that these meta-analyses have generally combined patients with a wide range of disorders, from hematopoietic cell transplantation to trauma to hemorrhagic stroke. Thus, a significant benefit in one clinical setting could be diluted by a lack of benefit in another.

Similarly, an increased risk of thrombosis in one patient group (eg, hemorrhagic stroke in older patients with vascular disease) could be hidden by the absence of increased risk in another group (eg, trauma in young previously healthy individuals). (See 'Thromboembolic complications' below.)

Cost may also be a consideration, since these products are very expensive, costing thousands of dollars for a single dose. However, in some settings, the drug may be cost-effective if it prevents serious bleeding or bleeding complications. Many institutions have policies or checks on use to reduce the likelihood of inappropriate administration [52-54].

A retrospective analysis of data from the Premier Perspectives database, which contains discharge records from a sample of academic and non-academic United States hospitals, indicated that in 2008, 97 percent of 18,311 in-hospital uses of rFVIIa were for off-label indications [55]. Another database study of 701 indicated that 92 percent of uses of rFVIIa were off-label [56].

Choosing the best dose, frequency, and duration of therapy — The optimal dose, frequency, and duration of rFVIIa is not known for any specific individual with a specific underlying condition, and clinical practices vary widely [52,57-59]. Determining the best dose is especially challenging since there is no laboratory test that correlates well with clinical efficacy. (See 'Monitoring efficacy' below.)

General advice about dosing includes the following, with specific dose recommendations in the sections below:

●Dosing should be proportionate to the degree of hemostatic impairment (eg, use a lower initial dose for epistaxis and a higher initial dose for central nervous system (CNS) bleeding; use more frequent dosing for serious bleeding and less frequent dosing for minor bleeding).

●The underlying hemostatic defect also influences dose and dosing interval. As examples, (dosing for NovoSeven) [41]:

•Glanzmann thrombasthenia (and possibly other platelet disorders) - 90 mcg/kg every two to six hours as needed [60]. (See 'Glanzmann thrombasthenia' below.)

•Congenital factor VII deficiency – 15 to 30 mcg/kg every four to six hours. (See 'Factor VII deficiency' below.)

Dosing for off-label indications such as trauma-associated coagulopathy is discussed below. (See 'Off-label uses' below.)

Dosing for hemophilia with inhibitors and acquired hemophilia is discussed separately. (See "Treatment of bleeding and perioperative management in hemophilia A and B", section on 'Bypassing products (rFVIIa products or FEIBA)' and "Acquired hemophilia A (and other acquired coagulation factor inhibitors)", section on 'Control bleeding'.)

●The dose is always rounded up to use complete vials (without discarding any of the product), except for small children.

●The duration of therapy is highly individualized based on the judgment of the treating clinician, balancing the likelihood of rebleeding with concerns about thromboembolic complications. We generally continue administration until we are certain that bleeding has stopped but generally do not give additional doses beyond cessation of bleeding, in order to minimize the potential risk of thrombosis. (See 'Thromboembolic complications' below.)

●Therapy should be discontinued if it is ineffective. However, data from observational studies suggest that up to two additional doses may be given (if tolerated) before deciding that treatment has failed [58].

Monitoring efficacy — There is no laboratory test that reflects clinical efficacy of rFVIIa. The reduction of the prothrombin time (PT) caused by rFVIIa does not reflect in vivo hemostasis.

Thus, monitoring must be based on the patient's clinical status and clinical assessment of bleeding.

A test that could predict the hemostatic efficacy of rFVIIa would be useful. An ideal test would likely include the patient's plasma and platelets, since rFVIIa is thought to interact with endogenous clotting factors on the surface of endogenous platelets. Whole blood coagulation tests such as thromboelastography (TEG) or rotational thromboelastometry (ROTEM) have been used in some cases [61-63]. However, these tests have not been validated for predicting efficacy or effects on bleeding risk.

Pediatric considerations — The use of rFVIIa in children has also been expanding. Both NovoSeven and Sevenfact are approved for children [41,42].

In general, we would use rFVIIa in children with similar dosing, frequency, and duration as adults. Children are probably less likely than adults to have thrombotic complications of rFVIIa, possibly because they are less likely to have underlying prothrombotic conditions, particularly arteriosclerotic vascular disease.

Thus, the risk-benefit calculation in children may favor a lower threshold for using rFVIIa in the setting of bleeding or major surgery for one of the approved or off-label conditions described below.

Evidence for the efficacy and safety of rFVIIa in children come from retrospective cohorts and case series:

●A retrospective analysis of administrative claims data that included 3764 children described a dramatic increase in use of rFVIIa over the years following licensing of the product [64]. Approximately 40 percent of the off-label use was in infants <1 year of age. Thrombosis was reported in 11 percent of the off-label administrations. Approximately one-fourth of the entire cohort died, but the causes of death were not described.

●A registry study that included 388 children treated with rFVIIa identified cardiac surgery as the main setting for administration, used in approximately half [65]. The remaining uses included other surgeries, trauma, and hematology/oncology indications. Approximately two-thirds of the patients received a single dose (median, 114 mcg/kg; interquartile range, 90 to 181 mcg/kg). Treatment with rFVIIa was associated with fewer transfusions, and reported efficacy was 82 percent. The rate of thromboembolic complications was approximately 5 percent and did not appear to correlate with the dose level.

A review determined that there is no role for routine or prophylactic administration of rFVIIa in children undergoing cardiac surgery, but rFVIIa may be reasonable as rescue therapy for life-threatening bleeding that is refractory to conventional treatments [66]. (See 'Trauma or surgery' below.)

Obstetric considerations — Data on the use of rFVIIa in pregnancy are limited. One of the largest series involved 110 pregnant women treated with rFVIIa and found therapy to be effective in many of the cases [67]. Case reports and small series report efficacy in treating bleeding in pregnant women with a number of underlying indications:

●Factor VII deficiency [68]

●Acquired factor inhibitors [69]

●Inherited platelet disorders [70]

●Acquired coagulopathies [71]

●Postpartum hemorrhage [72,73]

Postpartum hemorrhage is reduced with the antifibrinolytic agent tranexamic acid, without increasing thrombotic events. However, rFVIIa may provide additional hemostatic support if needed. Details of decision-making and administration, along with supporting evidence from clinical studies, are presented separately. (See "Postpartum hemorrhage: Medical and minimally invasive management", section on 'Administer tranexamic acid' and "Postpartum hemorrhage: Medical and minimally invasive management", section on 'Recombinant factor VIIa'.)

In general, although pregnancy is considered a hypercoagulable state, the thrombotic risk of rFVIIa is likely to be lower in pregnancy compared with other prothrombotic conditions, likely because most pregnant women are healthy and do not have vascular disease or other underlying thrombotic conditions.

For pregnant women with Glanzmann thrombasthenia, the use of rFVIIa may provide control of bleeding without exposure to allogeneic platelets, allowing platelet transfusions to be "saved" for future life-threatening bleeding, similar to non-pregnant individuals. (See 'Glanzmann thrombasthenia' below.)

APPROVED INDICATIONS

Hemophilia A or B with inhibitors — Both NovoSeven and Sevenfact are approved for bleeding in individuals with hemophilia A or B with an inhibitor [41,42]. NovoSeven is also approved for perioperative use in hemophilia with inhibitors. Both products are approved for all age groups.

The use of these and other products, along with important information regarding coadministration with emicizumab, is presented in detail separately. (See "Treatment of bleeding and perioperative management in hemophilia A and B".)

Acquired hemophilia — NovoSeven is approved for bleeding or perioperative use in patients with acquired hemophilia (acquired inhibitors [autoantibodies] to endogenous clotting factors, usually factor VIII) [41]. Sevenfact does not carry this indication [74].

When used for acquired factor inhibitors, the recommended dose of rFVIIa is 70 to 90 mcg/kg, usually repeated every two hours depending on the severity of bleeding. For surgery, the first infusion is given immediately before the procedure. Subsequent infusions are given every two to three hours and continued at a tapering dose as needed to maintain hemostasis and allow wound healing.

Additional treatments for acquired hemophilia, inhibitor eradication, and expected natural history of these inhibitors are discussed separately. (See "Acquired hemophilia A (and other acquired coagulation factor inhibitors)", section on 'Management'.)

Factor VII deficiency — NovoSeven is approved for bleeding or perioperative use in patients with congenital factor VII deficiency [41]. Sevenfact does not carry this indication [74].

When used for factor VII replacement in individuals with factor VII deficiency, the dose of rFVIIa is 15 to 30 mcg/kg, which is much lower than for other indications. For surgery, the first dose is given immediately before the procedure, with subsequent doses every four to six hours until hemostasis is achieved. Some experts use a longer interval (eg, 12 hours). Dosing may need to be continued to ensure wound healing.

Alternatives to rFVIIa, including plasma-derived factor VII concentrates (available in some European countries) and other products, as well as additional information on dosing, evidence for efficacy, and complications, are discussed separately. (See "Rare inherited coagulation disorders", section on 'Factor VII deficiency (F7D)'.)

Glanzmann thrombasthenia — NovoSeven is approved for bleeding or perioperative use in patients with Glanzmann thrombasthenia (GT) [41]. Sevenfact does not carry this indication [74].

●Rationale for use – GT is an inherited platelet function disorder in which platelets lack the alpha_IIbbeta₃ integrin (previously called the glycoprotein [GP] IIb/IIIa receptor), which is the platelet receptor for fibrinogen. Platelet counts are normal. (See "Inherited platelet function disorders (IPFDs)", section on 'Glanzmann thrombasthenia'.)

When these individuals have bleeding or require surgery, platelet transfusions can be used. However, individuals with GT may develop alloantibodies against GPIIb/IIIa and/or against human leukocyte (HLA) antigens, rendering bleeding refractory to platelet transfusions. Thus, avoiding platelet transfusions for minor bleeding is desirable, allowing platelet transfusions to be reserved for life-threatening or major bleeding. (See "Refractoriness to platelet transfusion", section on 'Alloimmunization'.)

●Indications – There are two groups of GT patients for whom rFVIIa could be used to treat bleeding or to prevent surgical bleeding; an approach to selecting therapy is illustrated in the algorithm (algorithm 1):

•Individuals whose disease has become refractory to platelet transfusions due to development of platelet alloantibodies.

•Individuals who have not yet received platelet transfusions, to avoid development of alloantibodies. This approach would allow the individual to "save" platelet transfusions for truly life-threatening bleeding events in the future.

Platelet transfusions have generally been used for major bleeding, but rFVIIa should be considered for use alone, to prevent the development of alloantibodies against platelets, or in addition to platelet transfusion if platelet transfusions are ineffective in major bleeding.

While the FDA indication for NovoSeven is for bleeding refractory to platelet transfusion, a review of registry data and published reports documented frequent use, and similar efficacy, in patients without platelet refractoriness [75].

●Dosing – The recommended dose for NovoSeven is 90 mcg/kg, repeated at two- to six-hour intervals as needed during a surgical procedure, or at longer intervals (every two to six hours) postoperatively [57,60]. A similar approach is used for bleeding.

Higher doses, up to 270 mcg/kg, have been used in some cases when lower doses were ineffective [76].

The number of doses is individualized according to the severity of bleeding or procedure. A 2006 guideline from the United Kingdom Haemophilia Centre Doctors' Organisation (UKHCDO) suggests that dental procedures could be covered with three doses (one before and two after the procedure) and surgical procedures with one dose immediately before a procedure, followed by regular infusions every 90 minutes, and then extending the interval as long as there is a risk of rebleeding [57].

●Supporting evidence – Evidence for the efficacy of rFVIIa in treating bleeding or surgery in patients with GT comes from the GT registry (a prospective, web-based registry), a manufacturer-sponsored safety surveillance database, and various published case series. Some reports describe dramatic reductions in bleeding in individuals for whom other therapies were ineffective [58,61,77-80]. Antifibrinolytic agents have been co-administered with rFVIIa in some cases.

•Results from the GT registry were summarized in a 2021 report that evaluated 133 patients with GT who received rFVIIa (NovoSeven) for the treatment of 333 bleeding episodes and 157 surgical procedures [75]. Approximately one-third were <12 years of age on first admission. The median dose of rFVIIa was 90 mcg/kg per infusion, and generally one to three doses were given per admission (range, 1 to 10 doses). Some individuals also received other therapies including platelet transfusions and/or antifibrinolytic agents. Treatment was considered effective in 79 percent of bleeding episodes and 88 percent of surgeries. Therapy was generally well-tolerated, with one case of deep vein thrombosis (DVT) and one allergic reaction.

•An earlier series from 2004 evaluated 59 patients with GT (approximately one-fourth were children) who had 108 bleeding episodes and underwent 34 invasive procedures [58]. Approximately three-fourths were also treated with an antifibrinolytic agent. Treatment was considered effective in 75 percent of bleeding episodes and 94 percent of procedures. Therapy was most effective when the dose was at least 80 mcg/kg, when the dosing interval was within 25 hours, and when at least three doses were given before deciding that treatment was ineffective. Addition of an antifibrinolytic agent was stated not to have a major impact on efficacy, but confidence in this assessment is low since so many individuals received an antifibrinolytic agent. Gastrointestinal bleeding appeared to be less amenable to effective treatment with rFVIIa, although numbers were small. Adverse events included one deep vein thrombosis with pulmonary embolism and one episode of clotting in a ureter. Both patients with thrombotic complications were receiving continuous infusion of rFVIIa for several days in combination with antifibrinolytic agents.

●Complications – Risk factors for thromboembolic complications are discussed below. (See 'Thromboembolic complications' below.)

OFF-LABEL USES

General evidence for off-label use — Enthusiasm for rFVIIa as a potential broad-spectrum hemostatic agent was ignited in 1999 by the dramatic case of a patient who became coagulopathic following a gunshot wound [81]. All usual measures had failed to control his bleeding, and rFVIIa was given as a last resort. The nearly miraculous recovery of this patient triggered an interest in the use of high-dose rFVIIa therapy in patients with a wide range of bleeding conditions who could not be managed by conventional therapy.

Available studies have been explored in a number of systematic reviews and meta-analyses:

●A 2012 Cochrane review identified 29 trials (4290 participants) involving rFVIIa for surgical prophylaxis or treatment of surgical or other bleeding [82].

•Preoperative use of rFVIIa reduced red blood cell transfusions by approximately one unit but did not improve survival.

•Therapeutic use of rFVIIa to treat traumatic, surgical, or spontaneous bleeding was associated with a trend towards reduced transfusions and a trend towards improved survival, neither of which reached statistical significance. For survival, the risk ratio (RR) was 0.91 (95% CI 0.78 to 1.06).

●A 2011 systematic review evaluated 16 randomized trials and 10 observational studies involving use of rFVIIa for one of five clinical scenarios (intracranial hemorrhage, cardiac surgery, trauma, liver transplantation, and prostatectomy) [83]. There was no mortality reduction for any of the indications; other outcomes were not presented.

●A 2008 meta-analysis identified 22 randomized trials of rFVIIa in 3184 patients with various acquired bleeding episodes (trauma, surgery, intracerebral hemorrhage) [84]. Compared with placebo, rFVIIa was associated with lower likelihood of receiving a transfusion (odds ratio [OR] 0.54; 95% CI 0.34-0.86). There was a trend towards improved survival with rFVIIa that did not reach statistical significance (OR 0.88; 95% CI 0.71-1.09), with approximately 15 percent mortality in both groups.

Evidence for efficacy in specific underlying disorders is presented in the following sections.

Specific underlying disorders

Congenital bleeding disorders — The greatest amount of data in inherited bleeding disorders are for hemophilia and Glanzmann thrombasthenia (GT). (See "Treatment of bleeding and perioperative management in hemophilia A and B", section on 'Bypassing products (rFVIIa products or FEIBA)' and 'Glanzmann thrombasthenia' above.)

For other inherited bleeding disorders, evidence for the efficacy of rFVIIa is limited to case reports and small series. The role of rFVIIa in these disorders is discussed in separate topic reviews:

●Factor XI deficiency – (See "Factor XI (eleven) deficiency", section on 'Hemostatic therapies'.)

●von Willebrand disease – (See "von Willebrand disease (VWD): Treatment of major bleeding and major surgery".)

●Bernard Soulier syndrome – (See "Inherited platelet function disorders (IPFDs)", section on 'Specific disorders'.)

For platelet disorders other than Glanzmann thrombasthenia (GT), rFVIIa is generally reserved for individuals for whom platelet transfusion is ineffective (refractoriness to platelet transfusions). Use in GT is discussed above. (See 'Glanzmann thrombasthenia' above.)

Dosing in inherited platelet function disorders is in the range of 90 mcg/kg (range: 80 to 120 mcg/kg), given as a bolus dose. The typical interval is approximately every two hours (range: 1.5 to 2.5 hours). At least three doses should be administered to secure effective hemostasis.

Continuous infusion of rFVIIa following an initial bolus has been reported in congenital factor VII deficiency and hemophilia with inhibitors but is not included as a dosing strategy in the product information [85].

Evidence for efficacy is summarized above. (See 'General evidence for off-label use' above.)

Trauma or surgery — Following trauma or surgery, patients may develop a coagulopathy that manifests as uncontrollable microvascular bleeding. Mechanisms include dilutional coagulopathy, acidosis, hypothermia, release of tissue factor and other procoagulant substances, and activation of fibrinolysis. (See "Etiology and diagnosis of coagulopathy in trauma patients", section on 'Etiologies'.)

Trauma or surgical coagulopathy is generally managed by careful attention to controlling hypothermia and acidosis and with blood component transfusion. Transfusion therapy should be optimized to the extent possible before administering rFVIIa [86]. If transfusion is required, using a 1:1:1 (Fresh Frozen Plasma to platelets to red blood cells [RBCs]) approach is considered prudent, and tranexamic acid is associated with improved survival as discussed separately. (See "Etiology and diagnosis of coagulopathy in trauma patients".)

rFVIIa is generally reserved as salvage therapy to control microvascular bleeding when conventional therapy with transfusions and antifibrinolytic therapies has failed to control the bleeding. As discussed below, rFVIIa appears to reduce blood loss and transfusion requirements but does not impact mortality.

Another potential indication for rFVIIa is trauma or surgery in an individual who declines blood transfusions, such as a Jehovah Witness (JW), or an individual for whom compatible blood is not available. Several case reports have described efficacy during surgery in JWs [87-90]. Additional details of management are described separately. (See "Approach to the patient who declines blood transfusion".)

The optimal rFVIIa dose for trauma-associated or surgical bleeding is unknown; doses described in case reports and small series vary markedly, from 20 to 40 mcg/kg to up to 180 mcg/kg [46,91-96]. A reasonable initial dose is 30 to 50 mcg/kg, repeated approximately every 90 minutes until bleeding (or concern for serious bleeding) has stopped [97].

Evidence supporting a role for rFVIIa in traumatic or surgical bleeding comes from several sources:

●Trauma:

•General – A 2005 randomized trial involving 301 patients with blunt or penetrating trauma found that compared with transfusions, rFVIIa (administered after the eighth unit of RBCs at a dose of 200 mcg/kg, followed by a dose of 100 mcg/kg one and three hours later) was associated with a reduction in transfusions [47]. There was no mortality difference. A registry study involving 380 injured patients seen at trauma centers found the best response in those with a blood pH ≥7.2, systolic blood pressure >90 mmHg, and platelet count ≥100,000/microL [98].

•Traumatic brain injury (TBI) – A 2009 retrospective study of 179 patients with TBI found a reduced length of stay and reduced use of plasma transfusions for those admitted to the intensive care unit who had received rFVIIa [99]. Another retrospective study that evaluated 63 patients with TBI reported similar findings [100]. A third retrospective study involving 64 patients with TBI found increased mortality in those treated with rFVIIa [101].

●Surgery:

•General – A 2008 meta-analysis identified seven randomized trials (772 participants) that evaluated rFVIIa versus placebo, or different doses of rFVIIa, in patients undergoing major surgery [102]. There was no difference in mortality (2.8 versus 2.3 percent), but rFVIIa was associated with a decrease in the number of transfusions (56 versus 68 percent; OR 0.52; 95% CI 0.31-0.86). A subgroup analysis suggested the benefit was mostly associated with administration of higher doses of rFVIIa (>50 mcg/kg).

•Abdominal/urologic – A 2008 meta-analysis focused on abdominal and urologic surgery identified a number of case reports and small series but no randomized trials [103]. Among 50 patients in 10 case series, 39 (78 percent) had a reduction in bleeding with rFVIIa. In a trial that randomly assigned 64 individuals with severe pancreatitis to receive or not receive a preoperative dose of rFVIIa (40 mcg/kg), the rFVIIa group had shorter surgical time and fewer transfusions, but overall mortality was not affected [104].

•Cardiac – A 2011 meta-analysis of trials evaluating rFVIIa for cardiac surgery found reductions in blood loss and/or the number of transfusions, but the metrics differed and could not be directly compared [105]. There was a trend towards reduced surgical re-exploration that did not reach statistical significance (13 versus 42 percent; OR 0.27; 95% CI 0.04-1.9) and no difference in mortality (13 versus 12 percent). Another cardiac surgery analysis of observational data involving 503 individuals found a reduction in transfusions [106].

Several of these analyses also reported increased risks of thromboembolism, including strokes. (See 'Evidence from clinical trials on thromboembolic complications' below.)

It is important to keep in mind that trauma and surgical patients are both very heterogeneous groups with a variety of underlying conditions and comorbidities and a diverse range of bleeding manifestations. There is no blanket recommendation regarding indications, dosing, or risks of rFVIIa that can be made for all such individuals. Involvement of a practitioner with expertise in hemostatic disorders is advised.

Liver disease — Liver disease, especially cirrhosis, is associated with a complex coagulopathy that may increase risks of bleeding and thrombosis. (See "Hemostatic abnormalities in patients with liver disease".)

Individuals with liver disease often require invasive procedures, raising questions about whether to administer hemostatic products. However, a rebalancing of hemostasis occurs in liver failure and hemostatic agents are generally not recommended prior to procedures [107]. rFVIIa is not helpful in major surgery or gastrointestinal or variceal bleeding in patients with liver disease.

●Coagulopathy – A 2006 evidence review concluded that the role of rFVIIa was less well-established than other therapies [108]. A small study in children with liver failure and a study of rFVIIa to facilitate placement of an intracranial pressure monitor found rFVIIa to be beneficial in stopping bleeding [109,110].

●Liver resection – One observational study and one randomized trial in liver transplant or liver resection showed no benefit of rFVIIa [111,112].

●Gastrointestinal bleeding – A randomized trial involving 245 individuals with cirrhosis and gastrointestinal bleeding also failed to demonstrate a benefit of rFVIIa over standard treatment [113]. Portal pressure appears to be a more significant factor than coagulopathy in variceal bleeding [114].

Management of variceal and other gastrointestinal bleeding in patients with cirrhosis is discussed separately. (See "Gastrointestinal endoscopy in patients with disorders of hemostasis", section on 'Cirrhosis' and "Overview of the management of patients with variceal bleeding", section on 'Resuscitation and support' and "Acute liver failure in adults: Management and prognosis", section on 'Bleeding prevention'.)

CNS bleeding — Central nervous system (CNS) bleeding is especially concerning due to potentially catastrophic outcomes. The brain is rich in tissue factor (TF) that is not saturated with factor VIIa, suggesting that rFVIIa may be effective by a mechanism involving TF rather than platelets. However, the risk of thrombosis is also of greater concern with CNS bleeding.

Thus, if rFVIIa is used in CNS bleeding, low doses (20 to 40 mcg/kg) are appropriate.

●Spontaneous intracerebral hemorrhage (ICH) – For spontaneous ICH, rFVIIa may be used as an adjunct to other therapies, although the evidence for benefit in patient-important outcomes is weak. The FAST trial randomly assigned 841 adults with no prior bleeding disorder who had a spontaneous ICH to receive a high dose of rFVIIa (80 mcg/kg), a low dose of rFVIIa (20 mcg/kg), or placebo, within four hours of symptom onset [115]. There was a dose-response reduction in the magnitude of hematoma expansion at 24 hours (placebo, 26 percent; low-dose rFVIIa, 18 percent; high-dose rFVIIa, 11 percent). However, there was no effect on clinical outcomes including functional recovery or survival (approximately 80 percent in all groups). Thromboembolic events were similar in all three groups, but the high dose of rFVIIa was associated with an increase in arterial thrombosis (placebo, 4 percent; high-dose rFVIIa, 8 percent). (See 'Thromboembolic complications' below.)

The efficacy of rFVIIa (or lack thereof) in intracerebral hemorrhage and decision-making regarding use are discussed separately. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Other hemostatic strategies not recommended'.)

●Traumatic brain injury – Data are mixed on the benefit of rFVIIa for traumatic brain injury, as discussed above. (See 'Trauma or surgery' above.)

Management is discussed separately. (See "Management of acute moderate and severe traumatic brain injury".)

Pulmonary hemorrhage — Reports continue to emerge on the successful use of rFVIIa in patients with pulmonary hemorrhage (eg, diffuse alveolar hemorrhage or massive hemoptysis) following a variety of insults including pneumonia, hematopoietic cell transplantation, metastatic choriocarcinoma, and microscopic polyangiitis [116-121].

A 2017 retrospective review of very low birth weight premature infants who developed life-threatening hemorrhage while in the neonatal intensive care unit found that one dose of rFVIIa (50 mcg/kg) within 30 minutes of the onset of pulmonary hemorrhage was effective in stopping pulmonary hemorrhage and reducing the need for blood products, but the rates of intraventricular hemorrhage and mortality were not decreased [122]. (See "Overview of short-term complications in preterm infants", section on 'Respiratory complications'.)

rFVIIa has also been successfully administered for pulmonary hemorrhage by intrapulmonary instillation in children and adults [123,124].

Anticoagulant-associated bleeding — There are no high-quality studies showing efficacy of rFVIIa over specific reversal agents for anticoagulant-associated bleeding. Further, individuals receiving anticoagulants are assumed to have a baseline increased risk of thrombosis (for which the anticoagulant was prescribed).

Thus, anticoagulant-associated bleeding should be treated with therapy specific to the anticoagulant, as discussed in separate topic reviews:

●Warfarin – (See "Management of warfarin-associated bleeding or supratherapeutic INR", section on 'Treatment of bleeding' and "Reversal of anticoagulation in intracranial hemorrhage", section on 'Warfarin'.)

●Direct oral anticoagulants – (See "Direct oral anticoagulants (DOACs) and parenteral direct-acting anticoagulants: Dosing and adverse effects" and "Reversal of anticoagulation in intracranial hemorrhage", section on 'Dabigatran' and "Reversal of anticoagulation in intracranial hemorrhage", section on 'Apixaban, edoxaban, and rivaroxaban'.)

If nonspecific hemostatic agents are indicated in anticoagulant-associated bleeding, an antifibrinolytic agent is generally preferred.

If rFVIIa is administered, the apparent "correction" of the prothrombin time (PT) or international normalized ratio (INR) does not reflect reversal of the anticoagulant effect, as the PT does not fully reflect the primary mechanism of action of rFVIIa on the platelet surface [125-127]. (See 'Implications for clinical use' above.)

ADVERSE EVENTS — The overall safety record of rFVIIa is favorable. However, concerns persist about thromboembolic complications, including arterial and venous events, since this is an activated factor that at therapeutic doses has a concentration 1000-fold higher than endogenous factor VII [128].

Thromboembolic complications — The product information for rFVIIa includes a Boxed Warning regarding serious arterial and venous thromboembolic events, the importance of discussing these risks with patients, and the need to monitor for (and educate patients about) signs and symptoms of thrombosis [41,74].

Incidence and risk factors — The likelihood of thromboembolic complications with rFVIIa is challenging to determine and varies according to patient population. The challenge arises in part because many patients treated with rFVIIa have other risk factors for thromboembolism, including prior thrombotic events, trauma, liver disease, surgery, immobility, and vasculopathies [56]. Studies have reported a range of incidence from as low as 1 percent to as high as 10 percent.

The risk is greatest in individuals with the most and most severe underlying thrombotic risk factors, which include [129]:

●Older age

●Underlying vascular disease

●Direct vascular injury

The risk is also thought to correlate with higher doses of rFVIIa, but not all studies have observed a strong correlation between dose and thromboembolic risk, especially in individuals who do not have underlying thrombotic risk factors. As an example, doses of up to 346 mcg/kg have been well tolerated in individuals with hemophilia [130].

Clinicians should also be aware of more serious concerns when treating individuals with hemophilia A who are receiving rFVIIa with emicizumab, although studies that showed thrombotic microangiopathy (TMA) with emicizumab plus activated prothrombin complex concentrate (aPCC; factor eight inhibitor bypassing activity [FEIBA]) did not show the same complication with emicizumab plus rFVIIa. This subject is discussed in detail separately. (See "Treatment of bleeding and perioperative management in hemophilia A and B", section on 'Patients with hemophilia A receiving emicizumab'.)

Evidence from clinical trials on thromboembolic complications — The following studies illustrate the thrombotic risk in selected clinical settings. Importantly, the baseline risk of thrombosis in these populations in not always available from a comparison group.

●Mixed populations – One of the larger analyses reviewed thromboembolic events from 35 randomized trials (4468 participants) of rFVIIa [131]. Overall, thromboembolic events occurred in 401 (9 percent). Compared with placebo, rFVIIa did not increase the rate of venous thromboembolism (VTE; 5.7 versus 5.3 percent). However, the rate of arterial thromboembolism was greater with rFVIIa (5.5 versus 3.2 percent; p = 0.003). These were mostly myocardial infarctions, and there was a correlation with age, with the highest rate of arterial thromboembolism in individuals 75 years or older (10.8 percent). When arterial events were analyzed according to the underlying cause of bleeding (spontaneous central nervous system [CNS] bleeding, liver disease, cardiac surgery, trauma, traumatic brain injury, others), all showed a trend towards increased risk of arterial thromboembolism, but the difference only reached statistical significance for spontaneous CNS bleeding (odds ratio [OR] for increased risk of arterial thrombosis with rFVIIa, 1.67; 95% CI 1.03–2.69).

●Spontaneous intracerebral hemorrhage – Two randomized trials of patients with intracerebral hemorrhage comparing rFVIIa at various doses with placebo did not observe an increase in thromboembolic events [115,132]. However, the FAST trial (rFVIIa at 0, 20, or 80 mcg/kg) noted that compared with placebo, the 80 mcg/kg dose of rFVIIa was associated with an increase in arterial thromboembolism (9 versus 4 percent; p = 0.04) [115].

●Acquired coagulopathy – In a review of 13 clinical trials using rFVIIa for coagulopathy secondary to treatment with anticoagulants, cirrhosis, or severe traumatic injury, adverse thrombotic events were similar with placebo and rFVIIa (5.3 versus 6 percent) [133].

●Surgery

•Cardiac – A 2011 meta-analysis of six trials using rFVIIa during cardiac surgery (470 participants) found an increased risk of stroke (OR 3.69; 95% CI 1.10-12.38) [105].

•General – A 2008 meta-analysis of seven trials in major surgery (722 participants) did not find an increase in thromboembolic complications (7.1 versus 5.3 percent; OR 1.32; 95% CI 0.69-2.52) [102].

●Vascular trauma – A review of thrombotic complications of rFVIIa found a high incidence of arterial thrombosis in individuals undergoing treatment for vascular trauma [134].

Prevention — As noted above, the risk of arterial thrombosis with rFVIIa appears to be greatest when used in individuals with underlying vascular abnormalities and/or vascular trauma. In many areas, data supporting the use of rFVIIa are very limited. In individuals at high risk of thrombosis, the safety risks need to be carefully considered when the benefit is unknown.

Additive or synergistic effects of other hemostatic agents are also a consideration. However, there are no data to support an increased risk of thrombosis when rFVIIa is used together with tranexamic acid.

Inhibitors — Inhibitors (neutralizing alloantibodies following exposure to rFVIIa that can react with endogenous factor VII or infused-factor rFVIIa) have been reported in individuals with severe congenital factor VII deficiency (factor VII activity level <10 percent).

●A case series that included 91 individuals with congenital factor VII deficiency (factor VII activity <10 percent) who were treated with rFVIIa reported inhibitor development in eight (9 percent) [135]. All inhibitors developed in childhood (age range, 3 months to 8 years). Patients were subsequently treated with high-dose rFVIIa or FEIBA (an aPCC), but three died of bleeding complications.

●A second case series that included 50 individuals with congenital factor VII deficiency treated with rFVIIa noted inhibitor development in two (4 percent) [136]. One had a factor VII level <1 percent and developed an inhibitor during childhood; the other was diagnosed with factor VII deficiency at the age of 60 years (factor VII activity, 6 percent) and developed the inhibitor at age 66 years.

A 2015 review of safety data noted that inhibitors have not been reported in patients receiving rFVIIa for indications other than congenital factor VII deficiency [137].

Inhibitors should be suspected in individuals who develop refractoriness to rFVIIa therapy after previously having a good response, especially those with severe factor VII deficiency.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately.

●(See "Society guideline links: Hemophilia A and B".)

●(See "Society guideline links: von Willebrand disease".)

●(See "Society guideline links: Rare inherited bleeding disorders".)

●(See "Society guideline links: Acquired bleeding disorders".)

SUMMARY AND RECOMMENDATIONS

●Mechanism of action – Recombinant factor VIIa (rFVIIa) is an activated clotting factor originally developed to manage bleeding in persons with hemophilia and inhibitory antibodies. At the supraphysiologic levels used clinically, the major mechanism of action is thought to involve tissue factor-independent thrombin generation on activated platelets. (See 'Background and mechanism of action' above.)

●Available products – There are two rFVIIa products in the United States (NovoSeven, approved in 1999; Sevenfact, approved in 2020). They differ in approved indications, production, and half-life. Both are approved for bleeding and perioperative management in hemophilia with inhibitors (all ages). NovoSeven is also approved for acquired hemophilia, Glanzmann thrombasthenia (GT), and factor VII deficiency. (See 'Available products and their properties' above.)

●Indications and dosing considerations – Use of rFVIIa must balance efficacy with thromboembolic risk, both of which vary with the underlying disorder(s). Outside management of hemophilia, factor VII deficiency, coagulation factor inhibitors, and Glanzmann thrombasthenia, we generally consider rFVIIa to be appropriate for patients with life-threatening bleeding or major surgery, and for whom hemorrhage cannot be adequately managed by transfusion or other conventional therapies. Dosing depends on the indication and the site and severity of bleeding (lower dose for epistaxis versus higher dose for central nervous system (CNS) bleeding; more frequent dosing for serious bleeding). (See 'Overview of situations in which rFVIIa may be helpful' above and 'Choosing the best dose, frequency, and duration of therapy' above.)

●Use in specific conditions

•Glanzmann thrombasthenia – rFVIIa can be used for bleeding or surgery in individuals whose disease has become refractory to platelet transfusions and to avoid development of alloantibodies against platelets and "save" platelet transfusions for life-threatening bleeding (algorithm 1). A suggested dose for NovoSeven is 90 mcg/kg, repeated at two- to six-hour intervals as needed. (See 'Glanzmann thrombasthenia' above.)

•Hemophilia A or B – rFVIIa can be used as a bypassing agent, as discussed separately. (See "Treatment of bleeding and perioperative management in hemophilia A and B", section on 'Inhibitors'.)

•Acquired factor inhibitors – rFVIIa can be used to treat certain bleeding manifestations. (See 'Acquired hemophilia' above and "Acquired hemophilia A (and other acquired coagulation factor inhibitors)", section on 'Control bleeding'.)

•Hereditary factor VII deficiency – Dosing is much lower than in other disorders (typically 15 to 30 mcg/kg per dose). (See "Rare inherited coagulation disorders", section on 'Factor VII deficiency (F7D)'.)

•Off-label uses – Most uses of rFVIIa are off-label, including other congenital bleeding disorders, coagulopathy from trauma or surgery, liver disease, and CNS or pulmonary hemorrhage. (See 'Off-label uses' above.)

The optimal dose and interval for rFVIIa are not known for most of these indications; reasonable dosing includes:

-Surgery/trauma – Transfusion is optimized before administering rFVIIa in surgical patients. If rFVIIa is used, a reasonable initial dose is 30 to 50 mcg/kg, repeated approximately every 90 minutes until bleeding (or concern for serious bleeding) has stopped. (See 'Trauma or surgery' above.)

-Specific sites – (See 'Liver disease' above and 'CNS bleeding' above and 'Pulmonary hemorrhage' above.)

rFVIIa generally is not used for anticoagulant-associated bleeding. (See 'Anticoagulant-associated bleeding' above.)

●Monitoring – There is no laboratory test that reflects clinical efficacy of rFVIIa. Monitoring is based on the patient's clinical status and clinical assessment of bleeding. (See 'Monitoring efficacy' above.)

●Considerations in children and pregnancy – Data on rFVIIa in children and during pregnancy are limited but continue to expand. The risk of thromboembolic complications is likely to be lower in these populations. Indications and dosing are similar to other groups. (See 'Pediatric considerations' above and 'Obstetric considerations' above.)

●Adverse effects – The major complication of rFVIIa is thromboembolism (venous and arterial, including strokes). Many individuals treated with rFVIIa have a baseline increased thromboembolism risk due to older age, underlying vascular disease, and direct vascular injury. (See 'Adverse events' above.)