Article Text
Abstract
Introduction Recombinant activated factor VII (rFVIIa) has been used in the management of traumatic haemorrhage for a decade. Anecdotally, its use is decreasing. The aim of this study was to define the use of rFVIIa in UK deployed military hospitals over the last 5 years.
Methods A retrospective database review was performed, using the Joint Theatre Trauma Registry, for the period January 2006 to June 2011. Data collected included use of rFVIIa, injury severity score (ISS), survival and injury pattern. The temporal trend of rFVIIa use, taking into account the number of severely injured patients presenting during each time period, was then analysed.
Results During the period January 2006 to June 2011, 156 injured patients received rFVIIa. 146 of these (94%) had an ISS >15; there were 45 fatalities. The median ISS among the group receiving rFVIIa was 30, and 20 patients had an ISS in the range 60–75. There was a significant reduction in the use of rFVIIa in the second half of 2010 and the first half of 2011, compared with the previous 12-month period.
Conclusion The use of rFVIIa in UK deployed military hospitals has declined since 2010, which is likely due to a combination of factors, including a change in resuscitation practice in these units, and a change in emphasis of manufacturer's guidance.
- rFVIIa
- recombinant factor VIIa
- military
- multiple trauma
- trauma
- education
Statistics from Altmetric.com
Introduction
Factor VII (FVII) is a plasma protein that is produced in the liver, and is involved in the coagulation cascade. Recombinant activated FVII (rFVIIa) (Novoseven, Novo Nordisk, Maaloev, Denmark) was developed as a treatment for haemophilia patients with antibodies to FVIII and FIX in the 1980s.1 ,2 The ability of rFVIIa to enhance thrombin generation on the surface of activated platelets has resulted in its use in other situations where the production of a stable blood clot is essential, one of which is in trauma patients with uncontrolled haemorrhage. In theory, rFVIIa should only provoke formation of clot where there is exposure of tissue factor, for example in vessels damaged by trauma.
Since the first reports from the Israeli defence medical services of rFVIIa being used to control traumatic haemorrhage in 1999, its use in military patients has had a particular resonance.3 As rFVIIa emerged as a potential treatment for haemorrhage, the conflicts in Iraq and Afghanistan resulted in hundreds of patients with severe haemorrhage that required complex management, and many patients had ongoing haemorrhage despite resuscitation and surgery. The use of rFVIIa therefore flourished, although was the subject of some controversy as to its efficacy and side effect profile.
However, recently haemostatic resuscitation and the use of bespoke transfusion of blood and blood components have been introduced into practice in the deployed UK military medical setting, with the use of targeted blood product replacement guided by bedside thromboelastometry. At around the same time, the summary of product characteristics for rFVIIa were changed to reflect evidence of an increase in adverse events, in particular the incidence of arterial thrombosis.
Anecdotally, the use of rFVIIa has decreased. The aim of this study was to define the use of rFVIIa, to establish a trend in use since the change in licensing, manufacturer's advice and development of clinical practice.
Methods
Data on severely injured patients (defined by activation of a trauma team, and including UK military, coalition forces, detainees and local civilians) treated by UK Defence Medical Services in deployed medical facilities are collected by the deployed clinical team and returned to the UK Joint Theatre Trauma Registry (JTTR) at the Royal Centre for Defence Medicine in Birmingham. Data are prospectively collected from clinical notes, trauma charts and in the case of death in UK servicemen, post mortem findings. The JTTR holds continuous data on this cohort from 2003, coinciding with the start of hostilities in Iraq. The entry criteria were expanded in 2007 to include all trauma patients who returned to the Royal Centre for Defence Medicine for definitive treatment, irrespective of whether a trauma team response was mandated.
A database review was undertaken to produce a dataset of patients presenting to a deployed military medical facility who received rFVIIa as part of their management. Until 2006, rFVIIa was not recorded as a separate treatment intervention, and therefore the search was conducted from the time it started to be recorded. Data collected included use of rFVIIa, injury severity score (ISS) 2005 (military) version, survival and injury pattern. Within the JTTR, the injuries are identified by Abbreviated Injury Scale (AIS) codes,4 reflecting increasing severity of injury for a given body region. The highest injury code for each patient was also recorded. To produce a temporal trend in use of rFVIIa and to take into account the variation in numbers of patients being treated in any given time period, the proportion of severely injured (ISS>15) patients, with vital signs present on arrival at hospital, receiving rFVIIa was calculated. The temporal trend was then analysed using a test for equality of different proportions over time to see if there was a significant difference in use.
Results
During the period January 2006 to June 2011, 5170 injured patients presented to deployed medical facilities, of whom 156 received rFVIIa. One hundred forty-six of these (94%) had an ISS>15; there were 45 fatalities. The median ISS among the group receiving rFVIIa was 30 (IQR 24–41). Twenty patients had an ISS in the range 60–75. rFVIIa was given to three patients who were classified as killed in action (KIA), or in whom vital signs were absent on arrival in the medical treatment facility. The trend of rFVIIa use during this period is shown in figure 1. There was a statistically significant reduction in the use of rFVIIa in the second half of 2010 and first half of 2011, compared with the previous 12-month period (difference in proportion treated −15.8%, 95% CI −20.0% to −11.4%, p<0.0001). There was no difference in mortality.
The mechanism of injury of those receiving rFVIIa is shown in figure 2. Lower extremity injury was the most common body region to have the highest AIS score, reflecting the fact that lower extremity amputation is the signature injury caused by the improvised explosive device (IED).
Discussion
The use of rFVIIa in military medical treatment facilities has reduced significantly since early 2010. This may be multifactorial, as there have been several factors that may have influenced its use over the last couple of years.
Activated FVII (FVIIa) is present in the normal circulation, and represents about 1% of the total plasma FVII.5 Following initial development from purification from plasma in the 1980s, recombinant technology was then employed to produce substantial quantities for therapeutic use.6
In patients with normal coagulation, rFVIIa binds to tissue factor exposed at the site of vessel injury, to initiate coagulation.7 It also has an effect independent of tissue factor, thought to be due to its binding to activated platelets at the site of vessel injury, again resulting in activation of coagulation.8 ,9 In the absence of other clotting factors, such as FVIII or FIX (haemophilia A and B, respectively), or acquired factor deficiency (eg, following trauma and subsequent resuscitation), rFVIIa binds to the activated platelet surface independent of the presence of FVIII or FIX, causing activation of FX. Giving a patient rFVIIa therefore results in a thrombin ‘boost’, and enhanced clot production, independent of the presence of other clotting factors. However, to work effectively it needs adequate functioning platelets.
The fact that rFVIIa enhances coagulation has led to its use in the management of haemorrhage in a wide range of circumstances, including control of bleeding from thombocytopathy,10 during cardiac and thoracic surgery,11 ,12 following postpartum haemorrhage,13 ,14 and following spontaneous intracerebral haemorrhage.15 ,16 Its use in the management of traumatic haemorrhage was first described in 1999,3 and it has since been the subject of three randomised controlled trials in trauma patients (table 1). No benefit in terms of survival has been shown in these studies, although there may be a potential benefit in reduction of blood product transfusion and a reduction in the incidence of ARDS.19
rFVIIa is currently licensed within the European Union for treatment of congenital haemophilia with inhibitors, acquired haemophilia, Glanzmann's thrombasthenia, and congenital FVII deficiency. However, in one study of 12 644 hospitalisations where rFVIIa was used, 97% were for reasons outside these licensed indications.20
The UK Clinical Guidelines for Operations21 include guidance on when to use rFVIIa, and in particular, for use in life-threatening haemorrhage. These guidelines suggest that rFVIIa should not be used in cases where the patient is expected to be unsalvageable despite administration of rFVIIa. During the period studied, rFVIIa was given to three patients who were KIA, in other words had absent vital signs on arrival at the hospital. These were all double or triple amputees, who arrived at the medical facility with absent vital signs but who received rFVIIa as part of their attempted resuscitation.
Several factors may have influenced the use of rFVIIa in deployed military hospitals over the last couple of years. Since 2006, damage control (haemostatic) resuscitation has been an integral part of deployed clinical practice and includes the initial use of equal volumes of blood and plasma in trauma resuscitation, with early replacement of platelets, to pre-emptively manage coagulopathy.22 In practice this means initial ratios of packed red blood cells (RBCs) to thawed plasma of 1:1 from the initial ‘shock pack’, and early administration of platelets and cryoprecipitate. The development of massive transfusion protocols, and protocols for the management of massive haemorrhage, has influenced resuscitation practice in civilian as well as military hospitals, and has been an integral part of haemostatic resuscitation.23 In a 2-year period from 2008 to 2010 one analysis found that 11% of military patients received massive transfusion (defined as >10 units packed RBCs in 24 h),24 similar to published figures from the US experience.25 The optimal ratio of plasma to RBCs is still the subject of some debate, with recent evidence pointing towards an optimal ratio of between 1:2 and 3:4, rather than 1:1.26
The main factor affecting rFVIIa use is likely to be the introduction in 2009 of target based resuscitation with near patient testing, as part of the development of damage control resuscitation practice. At around this time, rotational thromboelastometry (ROTEM) was introduced into clinical practice at Camp Bastion27 and allowed bespoke replacement of blood and blood products depending on the specific deficiencies according to the ROTEM result. This has allowed therapy targeted at the source of the coagulopathy, in contrast to the blanket approach of giving rFVIIa and generating a thrombin burst, with its potential for systemic adverse events. Rather than being forced to avoid the use because of the risk of adverse events, the current practice in Bastion has therefore allowed targeted resuscitation without the need for rFVIIa in the frequency previously recorded. However, it is still used for certain indications, such as continuing haemorrhage despite adequate replacement of blood products, clotting factors and platelets. Anecdotally, during the author's period in Camp Bastion between August and October 2011, it was used for one patient who was a triple amputee with a large area of soft tissue injury to both buttocks, who was continuing to ooze blood from these sites despite appropriate resuscitation, and another patient who sustained a gun shot wound to the chest, and had uncontrolled haemorrhage coming up his endotracheal tube. Both patients survived to aeromedical evacuation to their host country.
The other factor that may have affected rFVIIa use is that the manufacturer's recommendations have been reviewed and revised in light of the adverse event profile of the drug. It has now been emphasised in the summary of product characteristics that it should only be used within its licensed indications.
It is possible that rFVIIa use will in the future be targeted at specific injury patterns that are not amenable to surgical haemorrhage control, and those patients who are resistant to conventional haemostatic resuscitation. An example of this may be the use of rFVIIa in blast lung injury. The unlicensed use of rFVIIa was specifically endorsed for use to mitigate the haemorrhagic phase of blast lung injury by the UK Advisory Group on Military Medicine in August 2010; indeed, several patients during the peak of use of rFVIIa in 2009 were administered the drug for this indication.
Limitations of this study are inherent from the method of data collection that makes up the JTTR, which only captures those patients who have data submitted from the field hospitals, or who have postmortem data from repatriation and coroners' postmortem findings in the UK. For patients who are local civilians, coalition forces, or other non-UK patients who die prior to arrival at a medical facility, no data is collected other than a body map of injuries. Detailed AIS scoring is therefore not possible on these patients. To define an appropriate denominator in order to assess a trend, patients who were KIA were excluded (eg, in months where a multiple casualty incident resulted in numerous KIA patients with ISS>15, this would have skewed the results). However, three patients were given rFVIIa during their attempted resuscitation despite being classified as KIA according to the strict database criteria.
Conclusion
The use of rFVIIa in UK deployed military hospitals has declined since 2010, which is likely due to a combination of factors, including a change in resuscitation practice in these units, and a change in emphasis of manufacturer's guidance.
Acknowledgments
The author thanks Paul Newell of the Centre for Health & Environmental Statistics, Plymouth University, for his advice on the statistical analysis of this data.
References
Footnotes
-
Competing interests None.
-
Provenance and peer review Not commissioned; externally peer reviewed.