Practical Transfusion Medicine 4th Ed.

37. Pharmacological agents and recombinant activated factor VIIa

Beverley J. Hunt1 & Simon J. Stanworth2

1Kings College, London, UK and Departments of Haematology, Pathology and Rheumatology, Guy's and St Thomas' NHS Foundation Trust, London, UK

2NHS Blood and Transplant, John Radcliffe Hospital, Oxford, UK


There continues to be interest in the use of pharmacological agents to reduce bleeding, in light of the concerns about blood safety and new data on the efficacy and safety of tranexamic acid. Pharmacological agents are used in two ways: either to prevent excessive bleeding or to treat established bleeding. The agents used can be broadly classified into four groups: antifibrinolytics, topical sealants, desmopressin and the recombinant prohaemostatic factors such as recombinant activated factor VIIa (rFVIIa).


These include the lysine analogues, tranexamic acid (TA) and epsilon-amino caproic acid (EACA), which are competitive inhibitors of plasminogen binding to fibrin and endothelial receptors, and aprotinin, a (related) serine protease inhibitor that inhibits a number of haemostatic enzymes but principally has a powerful direct antiplasmin effect. In vitro TA has approximately ten times the antifibrinolytic activity of EACA and is therefore assumed to be a more potent antihaemorrhagic agent. A Cochrane systematic review has shown that TA and other lysine analogues are efficacious in reducing bleeding and blood usage perioperatively, but the publication of the Clinical Randomization of an Antifibrinolytic in Significant Haemorrhage 2 (CRASH-2) trial has pushed TA to the fore, as an efficacious, safe and cost-effective tool in reducing mortality in bleeding trauma patients.

Tranexamic acid in traumatic bleeding

The CRASH-2 study was a randomized controlled trial (RCT) of TA versus placebo in the management of bleeding after trauma recruited 20 000 patients worldwide [1]. The primary outcome was death in hospital within four weeks of injury. All-cause mortality was reduced significantly with tranexamic acid (1463 (14.5%) tranexamic acid group versus 1613 (16.0%) placebo group; relative risk, 0.91; 95% CI, 0.85–0.97; p = 0.0035). The risk of death due to bleeding was significantly reduced by 9% (489 (4.9%) versus 574 (5.7%); relative risk, 0.85; 95% CI, 0.76–0.96; p = 0.0077). Not only was the drug shown to be efficacious in reducing death but it was also shown to be safe as there were no adverse events and, importantly for a drug that affects haemostasis, there were no increased thrombotic events; indeed there was a trend to a lower rate of arterial events in those receiving TA. Further analysis of the data showed that benefit was greatest the earlier that TA was given after injury and that there was a possibility of negative benefit given after 3–6 hours from injury [2]. This has resulted in a worldwide change in the practical management of massive bleeding; for example, there is a National Health Service England programme of implementation to ensure that TA is given by paramedics and ambulance staff on site prior to hospital injury.

The cost effectiveness of using TA in trauma has been calculated in three countries [3]: Tanzania as an example of a low-income country, India as a middle-income country and the UK as a high-income country. The cost of giving TA to 1000 patients was $17 483 in Tanzania, $19 550 in India and $30 830 in the UK. The estimated incremental cost per life year gained of administering TA is $48, $66 and $64 in Tanzania, India and the UK respectively, making it highly cost effective. The WHO has recently classed it as an essential drug.

Interestingly CRASH-2 showed no reduction in the use of blood components in those treated with TA. A recently published study of the use of TA in a military study has suggested that this is due to the higher survival rate with TA: living patients will require further blood components. Indeed, in the military study the group receiving TA had a 24% survival rate and required more blood than the control group.

Perioperative use of the lysine analogues

A Cochrane systematic review [4] analysed data from over 211 RCTs of the perioperative use of antifibrinolytics in over 20 000 participants. It suggested that TA and EACA were nonstatistically slightly less efficacious in reducing the need for transfusion and the need for re-operation than aprotinin (as expected from their lower Ki against plasmin) (Table 37.1).

Table 37.1 The summary statistics from a Cochrane systematic review of antifibrinolytic use for minimizing perioperative allogeneic blood transfusion. Reproduced from Guerriero et al. [4].


Risk reduction in the use of red cell transfusion (85% CI)

Risk reduction in the need for re-operation for bleeding (95% CI)


34%, i.e. RR = 0.66 (0.61–0.71)

0.48 (0.35–0.68)

Tranexamic acid

0.61 (0.54–0.69)

0.67 (0.41–1.09)


0.75 (0.58–0.96)

0.35 (0.11–1.17)

CI, confidence interval; EACA, epsilon aminocaproic acid.

A continuous infusion is given perioperatively. The dose of TA that has been used is variable, 2.5–100 mg/kg over 20 minutes preoperatively followed by 0.25–4 mg/kg/h delivered over 1–12 hours and 1 mg/kg for 10 hours. Antifibrinolytics can also be given to treat established fibrinolysis: the recommended dose for TA is up to 1–2 g by slow IV infusion.

Obstetric haemorrhage

The CRASH-2 clinical trialists have now embarked on an RCT of TA in obstetric haemorrhage, aiming to randomize 14 000 women over 5 years to tranexamic acid of 1 gm followed by another 1 gm for continued bleeding versus placebo [5].


The original and licensed regimen of aprotinin for use in high risk cardiac surgery [6,7] was for 2M Kallikrein inhibitory units (KIU) to the patient, 2M KIU to the cardiopulmonary (CPB) circuit and 50 000 KIU per hour during CPB. This reduced postoperative drainage loss by 81% and total haemoglobin loss by 89%. Since aprotinin is a bovine protein and thus can provoke an immunological reaction, a test dose is required. Aprotinin can also be used in established fibrinolytic bleeding; 500 000 KIU intravenous (IV) is a good antiplasmin dose.

Aprotinin was shown to reduce perioperative bleeding in over 80 randomized controlled studies. Indeed, it has been used as an example by Sir Iain Chalmers as an example of the need for systematic reviews [8]; how was it ethical to randomize patients to placebo in the later trials when so much data was already showing that it reduced blood loss significantly? In fact, worse was to come. The error of using a reduction in blood loss as a surrogate marker of reduction in premature mortality was shown. For several large open studies of the use of antifibrinolytics in cardiac surgery it was suggested that although it reduced bleeding, it was associated with increased risk of death and renal dysfunction compared with other antifibrinolytics [9]. The later BART study, an RCT comparing aprotinin versus EACA versus TA was halted by the data monitoring committee due to concerns about the death rate in patients receiving aprotinin [10]. A total of 74 patients (9.5%) in the aprotinin group had massive bleeding, as compared with 93 (12.1%) in the TA group and 94 (12.1%) in the EACA group (relative risk in the aprotinin group for both comparisons, 0.79; 95% confidence interval (CI), 0.59–1.05). At 30 days the rate of death from any cause was 6.0% in the aprotinin group, as compared to 3.9% in the TA group (relative risk, 1.55; 95% CI, 0.99–2.42) and 4.0 in the EACA group (relative risk, 1.52; 95% CI, 0.98–2.36). The relative risk of death in the aprotinin group as compared with that in both groups receiving lysine analogues was 1.53 (95% CI, 1.06–2.22). There was voluntary suspension of marketing aprotinin. Subsequently the use of aprotinin has declined and TA has largely replaced it.

Mechanism of action of antifibrinolytics

TA and EACA are synthetic lysine analogues that bind to the lysine binding sites on fibrinogen and other plasminogen receptors present on white cells and endothelium that would normally bind plasminogen, i.e. a competitive inhibitor.

Aprotinin is a basic serine protease inhibitor extracted from bovine lung. In high doses (150–200 KIU), it inhibits kallikrein; the licensed regimen achieves blood levels of about 200 KIU/mL [11,12]. However, even in lower concentrations, aprotinin is a powerful inhibitor of plasmin, which appears to be the main mechanism for its effect on bleeding; its molar potency in vitro is 100 and 1000 times that of tranexamic acid (TA) and epsilon amino caproic acid (EACA).

Antifibrinolytics may also have a minor effect in preserving platelet membrane receptors, possibly by inhibiting plasmin-mediated degradation.

Monitoring the antikallikrein effect of aprotinin

Aprotinin by inhibiting kallikrein prolongs in vitro tests of the intrinsic system including the activated clotting time (ACT), which is used to monitor heparin during cardiopulmonary bypass. Kallikrein normally operates a positive feedback on the generation of factor XII. In order to allow for adequate levels of heparin, the ACT should be run greater than the normal level of 500 seconds, ideally at 750 seconds to compensate and allow for ‘normal’ heparin levels. The activator in the ACT has traditionally been celite, but kaolin has been used instead in some ACT tubes, for it is less affected by aprotinin and thus ACTs can be monitored in the normal way.


Desmopressin acetate (DDAVP) is a synthetic vasopressin analogue that is relatively devoid of vasoconstrictor activity. It increases the plasma concentrations and activity of von Willebrand factor (vWF) two- to fivefold by inducing the release of vWF from Weibel Palade bodies in the endothelium. It also stimulates the release of tissue plasminogen activator from the endothelium and promotes platelet activation. DDAVP shortens the bleeding time in patients with von Willebrand's disease, platelet function defects and uraemia and so is used for these indications.

Despite the success of early trials, a Cochrane systematic review of all 18 RCTs where DDAVP was given to reduce the use of allogeneic red cells concluded that there was no benefit from DDAVP in minimizing perioperative allogeneic red cell transfusion [13]. Side effects include flushing and an antidiuretic effect.

Topical sealants

Topical sealants can be used to stop oozing from small, sometimes inaccessible, blood vessels during surgery when conventional surgical techniques are not feasible. A number of sealants are licensed and used in surgery and trauma including:

·        liver and spleen lacerations;

·        dental extraction in patients with bleeding disorders;

·        gastric ulcers;

·        vascular grafts;

·        sealing of dural leaks; and

·        as an alternative to sutures.

Fibrin sealants mimic the final part of the coagulation cascade in that a source of thrombin is added to fibrinogen concentrates in the presence of calcium and a clot forms. They can be administered by a ‘gun’, which produces mixing of the reagents. Some sealants have two additional ingredients: factor XIII and aprotinin to stabilize the clot.

The initial source of thrombin was of bovine origin, which led to the development of a postoperative bleeding due to the formation of antibodies to bovine thrombin, which cross-react against human factor V, leading to acquired factor V deficiency [14].

Although they are derived from blood components, fibrin sealants have a lower risk of transmitting infection than donor blood. A Cochrane systematic review of their efficacy found a total of seven trials, including 388 patients, that showed a reduction of exposure to red cell transfusion by a relative 54%, but the trials were of poor methodological quality and larger more rigorous trials are needed [15].

Other topical sealants are available such as FloSeal, which is a thrombin-gelatin haemostatic matrix; Tachosil, a horse collagen sponge with human fibrinogen and thrombin; HemoStase, a purified plant polysaccharide; and CT3 surgical sealant, which is a novel absorbable polyethylene gyycol/collagen biopolymer sealant. There are inadequate data on reducing morbidity, mortality, length of stay and postoperative complication rate to know whether their use is efficacious and cost effective.

Recombinant activated factor VIIa

Recombinant activated factor VII (rFVIIa) was developed as a treatment for bleeding episodes in haemophiliac patients with inhibitors to factor VIII or IX. It is approved in Europe for this indication and for the management of:

·        haemophilia A or B with inhibitors;

·        acquired haemopilia;

·        congenital FVII deficiency; and

·        Glanzmann's thrombasthenia with refractoriness to platelet transfusion with antibodies to GPIIb/IIIa and/or HLA antigens.

However, rFVIIa has also been used widely as an ‘off label’ treatment in patients with platelet dysfunction, thrombocytopenia and massive transfusion after major surgery or trauma in patients without a pre-existing coagulopathy. Initially the body of evidence for these indications was disappointingly mainly from case reports and case series, but now data exists from 25 RCTs (see below).

Mechanism of action of pharmacological doses of rFVIIa

About 1% of circulating factor VII is in the activated form and the amount of rFVIIa required for bypass is larger than this. Certainly in haemophilia patients the doses required are much higher than those that generate a plasma concentration adequate for its binding to tissue factor. Disagreement revolves around the issue of whether rFVIIa has an effect independent of tissue factor. It has been demonstrated in vitro that rFVIIa is able to weakly bind activated platelets and cause activation of FX. The explanation of rFVIIa needing to bind to platelets may explain why rFVIIa is located only at the site of bleeding. Others hold the view that VIIa binds to tissue factor in the normal way. Whatever the mechanism, coagulation occurs locally at the site of bleeding without disseminated activation.

Hereditary clotting factor deficiencies

Haemophilia with an inhibitor or acquired haemophilia

Like the endogenous protein, rFVIIa has a short half-life, approximately 2.7 hours in adults, but the half-life in children and in bleeding haemophiliacs is shorter. The dosing interval in treating haemophiliac bleeding episodes is 2 hourly, lengthened up to 4 hourly later in the course of treatment. A loading bolus followed by a continuous infusion of rFVIIa is also used. Even though the recommended dosage is 90 μg/kg, it is clear that the optimal dose and dosing intervals of rFVIIa have not been established with certainty; higher doses up to 300 μg/kg have proved to be more clinically efficacious in some patients.

Whilst the prothrombin time (PT) and activated partial thromboplastin time (APTT) are shortened after treatment with pharmacological doses of rFVIIa, these are indirect correlates of its action. The measurement of FVII clotting activity (FVIIa:C) in the treatment of haemophilia-related bleeding has led to a recommendation of a minimum level of 6–10 IU/mL and peak levels of greater than 30–50 IU/mL when giving IV boluses. These levels appear to be associated with clinical improvement in haemostasis. The use of thromboelastography and thrombin generation has also been explored in trying to find an in vitro measure that correlates with clinical response.

Inherited factor VII deficiency

Bleeding episodes in patients with inherited factor VII deficiency have responded to lower doses of rFVIIa than required in haemophiliacs with inhibitors: doses ranging from 15 to 20 μg/kg every 2–3 hours until cessation of bleeding are recommended. Patients with factor VII deficiency are the only known patients to develop antifactor VIIa antibodies after treatment.

Other congenital bleeding disorders

There are anecdotal reports of the successful use of rFVIIa in bleeding in von Willebrand's disease. Some consider rFVIIa to be the agent of choice in patients with factor XI deficiency, with similar low doses as used in those with factor VII deficiency.

Inherited platelet defects

Patients with rare, congenital platelet defects have had successful treatment of bleeding episodes and undergone surgery safely with rFVIIa treatment. These include disorders such as Glanzmann thrombasthenia (abnormalities of the platelet fibrinogen receptor glycoprotein IIb/IIIa) and Bernard Soulier syndrome (lack of the glycoprotein Ib platelet receptor).

Platelet dysfunction occurs in uraemia and with aspirin, clopidogrel and glycoprotein (GP) IIb/IIIa inhibitors in acute coronary syndromes. rFVIIa has been reported to control bleeding anecdotally in these situations.

Off-licence use in patient groups other than haemophilia

Many bleeding patients without haemophilia have now been treated, off-licence, with rFVIIa [16,17]. The patients' settings are very diverse, including surgery (especially cardiac), gastrointestinal bleeding, liver dysfunction, intracranial haemorrhage and trauma, for example. Patients with liver dysfunction often have disproportionately low factor VII levels compared to the other vitamin K-dependent factors. rFVIIa normalizes the PT in liver disease with a single dose of 5–80 μg/kg, the dose depending on the patient. In trauma, following early reports of the use of rVIIa, there was an avalanche of reports of rFVIIa being used in uncontrolled bleeding after surgery and trauma. Much of these early patterns of use were driven by case reports and small case series, open to real problems of publication bias and confounding.

Randomized controlled trial evidence

Data from 25 RCTs enrolling around 3500 patients have now evaluated the use of rFVIIa as both prophylaxis to prevent bleeding (14 trials) or therapeutically to treat major bleeding (11 trials), in patients without haemophilia. This literature provides a more robust means of assessing the effectiveness and safety of rVIIA. When combined in meta-analysis [18], the trials showed modest reductions in total blood loss or red cell transfusion requirements (equivalent to less than one unit of red cell transfusion). However, the reductions were likely to be overestimated due to the limitations of the data. For other endpoints, including clinically relevant outcomes, there were no consistent indications of benefit and almost all of the findings in support of and against the effectiveness of rFVIIa could be due to chance (the exception was thromboembolic events, see below). Other limitations applied in many trials, e.g. uncertainty about the protocols for use of blood components.

A first randomized placebo-controlled trial of rFVIIa in blunt and penetrating trauma after patients had been transfused eight units of red cells suggested trends to requiring less red cells in those with penetrating injuries, without differences in thromboembolic events, but the findings were not replicated in a larger trial [19]. One issue for the trauma trials is the data, which show that rFVIIa is less effective in those who are acidotic or in severe shock. As for trauma, in the studies of patients with intracranial bleeding, although there were promising results in earlier therapeutic studies, the findings were not replicated in subsequent larger trials.

In both prophylactic and therapeutic groups of trials, there was an overall trend to increased thromboembolic events [18]. The forest plots for total arterial thromboembolic events are shown in Figure 37.1 and reach statististical significance. Thromboembolic disease is multifactorial and for many of the patients in the clinical settings of the included studies, a higher risk of thrombosis might be expected, for example, related to immobilization and stroke. Underestimation of the rates of thromboembolic events from the RCTs compared to current hospital practice is also likely, as a history of thrombosis or vaso-occlusive disease was a common criterion for exclusion in most of the included studies. Levi et al. published the rate of thrombotic events in all published RCTs of rVIIa and company data [20]. They also showed no increased rate of venous thromboembolism but significantly higher rates of arterial events; e.g. 2.9% of those who had rVIIa had coronary thrombosis compared with 1.1% of controls. Rates of arterial events were particularly high in those over 65 years (9% versus 3.6%, p = 0.003) and the rates were especially high in those over 75 years (10.8% versus 4.1%, p = 0.02). Concerns about the thrombotic risk, especially increased rate of arterial events associated with the off-licence use of rVIIa led to a Black Box warning from the FDA in the USA early in 2010.

Fig 37.1 Forest plot of arterial thromboembolic events in randomized trials of rVIIA. Reproduced from Simpson et al. [18].


Safety considerations

The mechanism of rFVIIa in initiating haemostasis led to concerns that widespread coagulation could be precipitated, particularly if tissue factor were expressed in atherosclerotic vessels; in which case, administration of rFVIIa could cause acute thrombosis. However, more than 7 000 000 doses of rFVIIa have been given to haemophiliacs with a 1% incidence of serious adverse events including myocardial infarction, stroke and venous thromboembolism. Moreover, a recently published meta-analysis of seven RCTs using rFVIIa in surgical procedures showed no increased risk of thromboembolism or mortality rates.


Pharmacological agents are now considered part of the management of major bleeding after trauma or surgery, but key questions pertaining to their safety and efficacy remain unanswered for many. Practically, TA and EACA have been shown to be efficacious and safe and should be used in preference to aprotinin.

The extent of rVIIa's efficacy, dosage and frequency of use remains unclear. Its use is limited by its high rate of arterial thrombosis, especially in those over 65, and its cost and lack of licence for many indications. It would seem prudent for each hospital to draw guidelines on the use of rFVIIa for ‘rescue therapy’ whereby it is used only when ‘best practice’ management of blood component therapy has failed.

Key points

1. Antifibrinolytics have been shown to reduce the use of allogeneic red cells during surgery. At the current time, TA and EACA are recommended over aprotinin in view of concerns about the safety of the latter.

2. TA reduces mortality in bleeding trauma patients and should be given as early as possible after injury to maximize its effect.

3. Topical sealants, although widely used, need more research to assess their full risk–benefit analysis.

4. Desmopressin has not been shown to be beneficial in reducing bleeding in surgical patients.

5. The efficacy of rFVIIa in off-licence ‘rescue therapy’ in bleeding patients is uncertain and is associated with an increased rate of arterial thrombosis, especially in the elderly. It might be considered as part of a local protocol after ‘best practice’ use of blood component therapy has failed.


1. Henry DA, Carless PA, Moxley AJ et al. Anti-fibrinolytic use for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev 2007; CD001886.

2. CRASH-2 trial collaborators; Shakur H, Roberts I, Bautista R, Caballero J, Coats T, Dewan Y, El-Sayed H, Gogichaishvili T, Gupta S, Herrera J, Hunt B, Iribhogbe P, Izurieta M, Khamis H, Komolafe E, Marrero MA, Mejía-Mantilla J, Miranda J, Morales C, Olaomi O, Olldashi F, Perel P, Peto R, Ramana PV, Ravi RR & Yutthakasemsunt S. Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet 2010, 3 July; 376(9734): 23–32. Epub 14 June 2010.

3. CRASH-2 collaborators; Roberts I, Shakur H, Afolabi A, Brohi K, Coats T, Dewan Y, Gando S, Guyatt G, Hunt BJ, Morales C, Perel P, Prieto-Merino D & Woolley T. The importance of early treatment with tranexamic acid in bleeding trauma patients: an exploratory analysis of the CRASH-2 randomised controlled trial. Lancet 2011, 26 March; 377(9771): 1096–1101, e1–2.

4. Guerriero C, Cairns J, Perel P, Shakur H & Roberts I. CRASH 2 trial collaborators. Cost-effectiveness analysis of administering tranexamic acid to bleeding trauma patients using evidence from the CRASH-2 trial. PLoS One 2011, 3 May; 6(5): e18987.

5. Shakur H, Elbourne D, Gülmezoglu M, Alfirevic Z, Ronsmans C, Allen E & Roberts I. The WOMAN Trial (World Maternal Antifibrinolytic Trial): tranexamic acid for the treatment of postpartum haemorrhage: an international randomised, double blind placebo controlled trial. Trials 2010, 16 April; 11: 40.

6. Royston D, Bidstrup BP, Taylor KM & Sapsford RN. Effect of aprotinin on need for blood transfusion after repeat open-heart surgery. Lancet 1987; 2: 1289–1291.

7. Karkouti K, Beattie WS, Dattilo KM et al. A propensity score case–control comparison of aprotinin and tranexamic acid in high transfusion – risk cardiac surgery. Transfusion 2006; 46: 327–338.

8. Chalmers I. The scandalous failure of scientists to cumulate scientifically. Abstract to paper presented at The Ninth World Congress on Health Information and Libraries, 20–23 September 2005, Salvador, Brazil. Available at: .php?lang=en&id=36 (cited 5 October 2005).

9. Mangano DT, Tudor IC & Dietzel C, for Multicentre Study of Perioperative Ischemia Group and Ischemia Research and Education Foundation. The risk associated with aprotinin in cardiac surgery. N Engl J Med 2006; 354: 353–365.

10. Fergusson DA, Hebert PC, Mazer CD et al. A comparison of aprotinin and lysine analogues in high-risk cardiac surgery. N Engl J Med 2008; 358: 2319–2331.

11. Segal H & Hunt BJ. Aprotinin: pharmacological reduction of perioperative bleeding. Lancet 2000; 355: 1289–1290.

12. Hunt BJ, Segal H & Yacoub M. Aprotinin and heparin monitoring during cardiopulmonary bypass. Circulation 1992; 86: 410–412.

13. Carless PA, Henry DA, Moxet AJ et al. Desmopressin for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev 2004;CD001884.

14. Banninger H, Hardegger T, Tobler A et al. Fibrin glue in surgery: frequent developments of inhibitors of bovine thrombin and human factor V. Br J Haematol 1993; 85: 528–532.

15. Carless PA, Henry DA & Anthony DM. Fibrin sealant use for minimising peri-operative allogeneic blood transfusion. Cochrane Database Syst Rev 2003; CD004171.

16. Martinowirz U, Kenet G, Lubetski A, Luboshitz J & Segal E. Recombinant activated factor VII for adjunctive hemorrhage control in trauma. J Trauma 2001; 51: 431–438.

17. Shao YF, Yang JM, Chau GY et al. Safety and haemostatic effect of recombinant activated factor VII in cirrhotic patients undergoing partial hepatectomy: a multicentre, randomized, double blind, placebo-controlled trial. Am J Surg 2006; 191: 245–249.

18. Simpson E, Lin Y, Stanworth S, Birchall J, Doree C & Hyde C. Recombinant factor VIIa for the prevention and treatment of bleeding in patients without haemophilia. Cochrane Database of Syst Rev 2011; 2: CD005011. DOI: 10.1002/14651858.CD005011.pub3.

19. Boffard KD, Bruno Riou B, Brian Warren B et al. Recombinant factor VIIa as adjunctive therapy for bleeding control in severely injured trauma patients: two parallel randomized, placebo-controlled, double-blind clinical trials. J Trauma 2005; 59: 8–15.

20. Levi M, Levy JH, Andersen HF & Truloff D. Safety of recombinant activated factor VII in randomized clinical trials. N Engl J Med 2010, 4 November; 363(19): 1791–800. Erratum in N Engl J Med 2011 17 November; 365(20): 1944.

Further reading

Freemantle N & Irs A. Observational evidence for determining drug safety. Br Med J 2008; 336: 627–628.

Dzik WH, Blajchman MA, Fergusson D, Hameed M, Henry B, Kirkpatrick AW, Korogyi T, Logsetty S, Skeate RC, Stanworth S, Macadams C & Muirhead B. Clinical review: Canadian National Advisory Committee on Blood and Blood Products – Massive Transfusion Consensus Conference 2011: report of the panel. Crit Care 2011, December 8; 15(6): 242.

Roberts HR, Monroe DM & White GC. The use of recombinant factor VIIa in the treatment of bleeding disorders. Blood 2004; 104: 3858–3864.

Rossaint R, Bouillon B, Cerny V, Coats TJ, Duranteau J, Fernandez-Mondejar E, Hunt BJ, Komadina R, Nardi G, Neugebauer E et al. Management of bleeding following major trauma: an updated European guideline. Crit Care 2010; 14: R52.