Gavin J. Murphy1, Nicola Curry2, Nishith Patel3 & Timothy S. Walsh4
1School of Cardiovascular Sciences, University of Leicester, Leicester, UK
2John Radcliffe Hospital, Oxford, UK
3School of Clinical Sciences, University of Bristol, Bristol, UK
4Edinburgh University and Edinburgh Royal Infirmary, Edinburgh, Scotland, UK
Good blood management emphasizes the importance of utilizing blood components as part of an overall treatment strategy that is focused on improving patient outcome. Acute haemorrhage and acute anaemia are common in surgical, obstetric and critical care patients. They are also prevalent in nonsurgical patients with upper gastrointestinal haemorrhage. These often critically ill patients are characterized by:
1. A high red cell transfusion requirement.
2. An association with urgent or emergency procedures or clinical events. Patients are usually cared for in highly monitored environments in which cointerventions with therapeutic adjuncts and the use of evidence-based protocols can reduce the use of conventional blood components.
3. Coagulopathy that requires management to assist correction of cardiovascular instability and anaemia.
This chapter reviews the evidence to guide blood management strategies in patients with critical illness or who are undergoing major surgery with an emphasis on those that have been shown to improve clinical outcomes. It also specifically considers changes in the management of massive haemorrhage/blood transfusion that have occurred in recent years, chiefly as a result of the lessons learned from recent armed conflicts, as this represents a clinical situation where appropriate blood management is a key determinant of patient outcomes, including survival.
Red cell transfusion
Anaemia and acute haemorrhage
In surgical and critically ill patients this accounts for almost 50% of all red cell utilization. In a UK study  the main users of allogeneic red cells were upper GI haemorrhage (13.8%), orthopaedic surgery (6.3%), trauma (5.9%), liver/GI surgery (5.5%) and cardiac surgery (5.2%). Over 10% of all red cell transfusions are administered in the intensive care unit (ICU) setting.
In acute haemorrhage, the therapeutic priority is to achieve source control; during this period the aim is to maintain adequate oxygen delivery to prevent tissue hypoxia and organ dysfunction using fluids, red cells and interventions to prevent or correct coagulopathy. Once haemorrhage has been stopped management is similar to that for the acutely anaemic patient. This is supported by prospective epidemiological studies in critical care patients where transfusion indicators, i.e. haemoglobin thresholds, are similar in bleeding and nonbleeding patients . Anaemia in the absence of haemorrhage occurs in surgical patients as a result of low preoperative red cell mass or haemodilution and this may account for the greater proportion of all red cell transfusions. For example, in cardiac surgery severe haemorrhage occurs in up to 15% of patients but red cell transfusion occurs in 50–95% of patients, depending on institutional transfusion practice .
Acute anaemia is also common in critical care where the aetiology is multifactorial and includes haemodilution, occult blood loss, therapeutic blood sampling and/or impaired haemopoiesis, which may be acute as a result of sepsis, for example, or chronic as a result of chronic renal or other systemic disease. Anaemia is strongly associated with adverse outcomes in the critically ill and despite the use of multiple interventions and therapeutic adjuncts such as avoidance of haemodilution, excessive therapeutic blood sampling or other modalities, as listed below, red cell transfusion is common. Up to 35–45% of patients receive a blood transfusion within 5 days of ICU admission, of which as many as 90% are administered to reverse anaemia .
Indications for red cell transfusion
Whereas red cell transfusion for acute haemorrhage in patients with incipient hypovolaemic shock is clearly life-saving the indications for red cell transfusion to reverse severe anaemia in the critically ill are poorly defined. Observational studies suggest a high likelihood of harm including infection, ischaemic and pulmonary morbidity and mortality (Figure 27.1) from excessive red cell transfusion. However, this evidence is subject to numerous biases including residual confounding from unmeasured variables, treatment and regression bias that arises due to more liberal administration of red cells to sicker patients and publication bias. Determining whether there is a causal effect of transfusion on adverse outcomes from these types of study is impossible.
Fig 27.1 Forest plot summarizing the odds ratios from landmark papers evaluating the relationship between red cell transfusion and 30 day mortality. Upper panel shows odds ratios from observational studies after adjustment for measured confounders. Lower panel shows odds ratios from RCTs that have compared liberal versus restrictive transfusion thresholds. An increasing odds ratio indicates increasing risk of death with transfusion.
Higher levels of evidence such as RCTs comparing restrictive versus liberal transfusion thresholds that result in equally matched patient groups exposed to different transfusion volumes do not show an increase in adverse effects attributable to red cell transfusions, although a recent meta-analysis of these RCTs performed in adult patients suggests that more restrictive transfusion practice may be associated with modest benefits in terms of reduced infection, although with no benefits in terms of reduced cardiac or other complications . This meta-analysis included mainly small, poorly reported studies, with lack of blinding and allocation concealment in many cases raising the possibility of detection and performance bias. The results of the analysis were also heavily influenced by a single large RCT, the Transfusion Requirements in Critical Care Study that was undertaken over a decade ago, and administered nonleucocyte-reduced red cells to patients with a relatively low incidence of cardiovascular disease, a principal determinant of both tolerance of anaemia and the risk of ischaemic complications. The difference in transfusion volume between treatment groups in these RCTs was often less than 2 units of blood, an amount that observational data suggests as having only very modest effects on clinical outcomes; even large RCTs may be inadequately powered to detect such an effect. However, the important finding of this analysis was that restrictive transfusion practice in patients with acute anaemia and without cardiovascular disease is safe using transfusion thresholds of 7 g/dL. Moreover, the use of more liberal transfusion thresholds have no clinical benefit, and in fact may have adverse effects. Safe restrictive thresholds in patients with cardiovascular disease remain to be defined.
Treatment adjuncts that reduce transfusion (also see Chapters 34 and 35)
The safety and efficacy of commonly used techniques or interventions that reduce transfusion exposure as summarized in a series of recent systematic reviews [5–12] are summarized in Figures 27.2 and 27.3.
Fig 27.2 Forest plot summarizing the effects of commonly used therapeutic adjuncts aimed at reducing bleeding and transfusion on allogeneic red cell exposure. Data derived from recently published meta-analyses [5–12] as indicated. Box size represents relative precision of the estimate as derived from the sample size and variance.
Fig 27.3 Forest plot summarizing the effects of commonly used therapeutic adjuncts aimed at reducing bleeding and transfusion on important clinical outcomes. Data derived from recently published meta-analyses [5–12] as indicated. Box size represents relative precision of the estimate as derived from the sample size and variance.
Autologous transfusion techniques
Preoperative autologous donation (PAD)
PAD involves the patient donating one or more units of his/her own blood preoperatively, often in conjunction with the administration of erythropoietin. This blood is held within the blood transfusion laboratory, where it is administered as required during the perioperative stay, as an alternative to allogeneic red cells. PAD is effective at reducing exposure to allogenic blood (Figure 27.2), but overall exposure to transfused red cells (both autologous and allogeneic) is increased, and PAD is not associated with improved clinical outcomes (Figure 27.3). Autologous red cells are presumably as susceptible as allogeneic red cells to storage-related changes that have been linked to transfusion-related morbidity, and where the local allogeneic blood supply is safe from infectious diseases PAD may not confer any overall clinical benefit. PAD may be advantageous in less developed healthcare systems where transmission of infection by transfusion remains an issue. However, PAD is restricted to patients scheduled for elective surgery, requires significant investment in infrastructure for the harvesting, testing and storage of autologous red cells in parallel to the systems in place for allogeneic blood and its adoption has not been widespread.
Acute normovolaemic haemodilution (ANH)
ANH involves removing blood from a patient, usually during induction of anaesthesia, replacing it with crystalloid or colloid fluid to maintain circulating volume and storing the blood for reinfusion during surgery as a response to blood loss, or at the end of surgery. Significant haemodilution reduces the red cell mass lost during surgery and replacement of losses with autologous blood has better homeostatic properties than colloid or crystalloid. ANH may also improve haemostasis by preventing consumption or loss of clotting factors during prolonged procedures or as a result of cardiopulmonary bypass and ANH has been shown to reduce bleeding rates. ANH also significantly reduces allogeneic red cell exposure and is inexpensive (Figure 27.2). ANH has not been shown to result in specific clinical benefits to patients beyond reducing transfusion exposure (Figure 27.3). The disadvantages of ANH relate principally to the safety of low haematocrits during surgery, which may increase the risk of neurological, myocardial and renal injury.
Mechanical cell salvage
Blood lost as a result of acute haemorrhage during major surgery can be collected (salvaged) using commercially available and widely used devices that wash the blood, removing plasma proteins, cell fragments and other contaminants of the surgical field and allowing reinfusion of washed autologous cells. A systematic review of 75 RCTs in orthopaedic (36 studies), cardiac (33 studies), and vascular (6 studies) surgery has demonstrated that this technique significantly reduces red cell exposure (Figure 27.2) and more importantly improves clinical outcomes including the risk of perioperative infection (Figure 27.3). The net cost benefit of cell salvage has been estimated to be between £112 and £359 per person . The administration of unwashed blood or blood harvested from postoperative losses was not found to be harmful overall. However, published guidelines do not recommend the use of unwashed (risk of excessive bleeding) or postoperative shed mediastinal fluid (risk of infection) in the setting of cardiac surgery .
Recombinant human erythropoietin (RhEpo)
RhEpo is commonly administered along with iron supplementation to reverse chronic anaemia preoperatively in surgical patients, where it has been shown to reduce transfusion exposure without apparent adverse effects (Figures 27.2and 27.3). A meta-analysis of RCTs evaluating the use of RhEpo and iron + RhEpo in critical care patients also shows a reduction in red cell exposure (odds ratio of transfusion 0.73, 95% confidence intervals (CI) 0.64–0.84) without apparent detriment (odds ratio of death 0.86, 95% CI 0.71–1.05) . The studies included in this meta-analysis were generally of poor quality and underpowered to detect important clinical outcomes. A more recent high quality RCT has shown an increased risk of developing thromboembolic complications attributable to RhEpo administration , although overall survival was improved in a subset of trauma patients in this study. The increase in thromboembolic complications is attributable to increased viscosity as well as a direct effect of RhEpo on platelet aggregation and is most evident in patients with existing renal and cardiovascular disease.
The lysine analogues tranexamic acid and epsilon-amino caproic acid (EACA) act by irreversibly binding to the active site of plasminogen, thereby inhibiting clot lysis. Tranexamic acid reduces transfusion exposure in a wide range of acute settings including cardiac, orthopaedic and liver surgery . More recently it has been shown to improve outcomes including survival if administered early in the management of acute blunt trauma . There is little consensus as to the most effective dose of these agents. Few adverse effects have been reported, but systematic reviews indicate that tranexamic acid does not improve clinical outcomes in cardiovascular surgery, unlike other clinical settings, suggesting a possible safety signal in patients at greatest risk of thromboembolic complications. Such an effect appears to be more pronounced with the serine protease inhibitor aprotinin, which acts as an antifibrinolytic as well as having a range of other anti-inflammatory and anti-apoptotic actions. Aprotinin has greater efficacy at reducing bleeding relative to tranexamic acid but data from RCTs suggests that this is also associated with an increased risk of adverse outcomes including mortality (Figure 27.3).
Desmopressin is a synthetic analogue of arginine vasopressin that induces the release of the contents of endothelial cell-associated Weibel–Palade bodies, including the von Willebrand factor. Its use is indicated in the management of patients with mild haemophilia and von Willebrand disease undergoing minor surgical procedures. The increase in Factor VIII and von Willebrand factor concentrations, as well as evidence of increased platelet aggregation in response to desmopressin, has led to its evaluation as a haemostatic agent in major surgery. A recent Cochrane review failed to demonstrate any significant reduction in transfusion exposure or improvement in clinical outcomes (Figures 27.2and 27.3). There were reductions in blood loss and the volume of red cells administered attributable to desmopressin use, but these were not deemed to be of clinical significance.
Recombinant activated Factor VII (rFVIIa)
This is a potent pharmacological prohaemostatic agent licensed for use in patients with haemophilia. This has led to the off-label use of rFVIIa for the treatment of severe coagulopathic bleeding in trauma and surgical settings as an adjunct to conventional non-RBC blood components. Its use is associated with a significant (68%) increased risk of major thrombotic complications, especially arterial thrombosis (Figure 27.3).
Coagulopathy is a poorly defined term; it may refer to severe impairment of blood coagulation in the setting of trauma or, alternatively, to the laboratory finding of abnormal screening tests of coagulation in a critical care patient. The lack of a clear definition of coagulopathy complicates epidemiological analyses and the development of accurate diagnostic tests and treatments. However, it remains a significant clinical problem and, depending on the definition, affects up to 30% of critically ill patients, 30% of trauma patients, 15% of cardiac surgery patients and 6% of those with acute upper GI haemorrhage. More detailed information on the underlying pathogenesis of acquired coagulopathy can be found in Chapter 25. Coagulopathic patients, whether or not they are actively bleeding, have a worse overall prognosis than similar patients without coagulopathy. This is attributable to the severity of the underlying illness and prior or ongoing significant haemorrhage and shock. It may also be attributable in part, however, to the adverse effects of prohaemostatic therapies; FFP and platelets are recognized causes of transfusion complications such as transfusion-related acute lung injury (TRALI), transfusion-associated circulatory overload (TACO), transfusion-associated dyspnoea (TAD) and transfusion-transmitted infection (TTI)  (see Chapters 6 to 16 for more details). Platelets have also been shown in some studies to increase the risk of stroke in patients with cardiovascular disease. These risks, although offset by the risks of ongoing bleeding in coagulopathic patients, may be clinically significant in those without coagulopathy, or when administered to those who are not actively bleeding.
Effective treatment of coagulopathy, particularly in a bleeding patient, requires accurate and timely diagnosis. The nature of coagulopathy is heterogeneous and is influenced by the patient group, e.g. severe trauma, liver surgery or cardiac surgery, the type of intervention, e.g. cardiopulmonary bypass or transplant surgery, and also by the blood management strategy adopted, e.g. the use of antifibrinolytics and non-red-cell blood components. Specific defects in the coagulation pathway are commonly not detected by standard coagulation screening tests that by taking as long as 65 minutes are often considered impractical in the setting of ongoing blood loss. Near patient testing is increasingly advocated in this setting.
The most widely used near patient testing devices include the Thromboelastogram (TEG®) or ROTEM® (also see Chapter 25). These are whole blood viscoelastic tests that evaluate the effects of coagulation factors, platelets and red cells on overall clotting potential. Both work along similar principles, whereby progressive clot formation in the presence of an activator is measured as impedance to a rotating pin within the clot. The resultant trace can then be used to infer information as to the activity of separate components of the clotting pathway, including the coagulation cascade, platelet function and lysis (Figure 27.4). These platforms, although user friendly and widely used, have limited sensitivity and specificity, and a recent systematic review has highlighted the lack of evidence of clinical benefit associated with their use . Near patient platelet function analysers and alternative laboratory assays such as thrombin generation testing have been shown to predict bleeding accurately and target therapy in small single-centre studies, but wider validation of these techniques is awaited.
Fig 27.4 ROTEM® Thromboelstograph trace and interpretation. Reproduced with permission from TEM International GmbH.
Without accurate diagnostic tests to identify specific defects in the coagulation pathway that are associated with adverse clinical outcomes, the management of coagulopathy is often empiric, nonspecific and based on the assumption that reversal of coagulopathy is beneficial. The clinical efficacy, safety and cost effectiveness of this approach is questionable, and this remains an important and underresourced area of research.
Acute haemorrhage during surgery is a common indication for therapeutic platelet use. For example, cardiac surgery utilizes over 17% of all platelet transfusions in the UK . Indications for and effective doses of platelets in the setting of acute haemorrhage are unclear and not supported by evidence. Observational studies report lower mortality rates in trauma patients receiving high dose platelet transfusion for major blood loss (Table 27.1), but there are no RCT data outside the haemato-oncology setting that can be used to guide practice (see Chapter 28 for further details).
Table 27.1 Efficacy of high dose platelet use in massive traumatic haemorrhage.
Thrombocytopenia in critically ill patients is a risk factor for major bleeding and death, and the prevalence of mild (<150 × 109/L) and moderate (<50 × 109/L) thrombocytopenia in adult ITU patients is reported at 40% and 8% respectively . There is little evidence from critical care studies that prophylactic correction of thrombocytopenia translates into a survival advantage, or indeed reproducibly raises platelet counts in critically ill patients. Thrombocytopenia increases the risk of haemorrhage during invasive procedures such as central line or spinal catheter insertion, and these are often performed using platelet transfusion ‘cover’. Consensus recommendations [19,20] for platelet administration during hae- morrhage and in the critically ill are summarized in Table 27.2. These thresholds for platelet transfusions are empirical, have been derived largely from studies in haemato-oncology patients and do not account for alterations in platelet function or clinical status, which limits their utility.
Table 27.2 Platelet thresholds for prophylactic and therapeutic platelet transfusion. Adapted from References  and .
Massive transfusion and multiple trauma or TBI
DIC and bleeding
Pre-invasive procedure, i.e. LP, CVC, epidural
Pre-surgery at high risk sites: i.e. brain/eye
TBI, traumatic brain injury; DIC, disseminated intravascular coagulation; LP, lumbar puncture; CVC, central venous catheter.
Nearly half of all FFP administered in the UK is to critically ill patients; 12% cardiac, 9% liver disease and liver transplant, 7% GI haemorrhage, 6% vascular surgery, 6% haematology, 3% trauma and 2% obstetrics . A recent UK study reported 13% of critically ill adult patients received FFP during an ITU admission . Half of these transfusions (48%) were for bleeding, while the remainder were for preprocedural prophylaxis (15%) or prophylaxis alone (36%). One-third were given to patients with normal PT values. Clinical efficacy of FFP has not been clearly demonstrated, however, either for treatment or prophylaxis. Indeed, it has been reported that standard FFP doses (12–15 mL/kg) are insufficient to significantly increase individual coagulation factor levels. A recent systematic review examining 80 RCTs highlighted that there are few well-supported indications for FFP administration  (see Table 27.3), but, despite this, numbers of FFP transfusions are increasing.
Table 27.3 Summary of FFP RCTs in critically ill patient groups. Adapted from Yang et al. .
Traditionally, fibrinogen is replaced during major blood loss or as part of the management for disseminated intravascular coagulation (DIC) once the Clauss fibrinogen value falls below 1 g/L. Fibrinogen is one of the earliest coagulation factors to fall in major bleeding and adequate, timely replacement of fibrinogen is hypothesized to result in improved haemorrhage control. New European trauma guidelines reflect this shift in practice and recommend 1.5 g/L as the transfusion trigger , although this recommendation is based on weak evidence. Cryoprecipitate is the first-line treatment in the UK for acquired hypofibrinogenaemia, and a standard adult dose (2 pools) raises the plasma fibrinogen level by 1 g/L. There are no clinical data to confirm effectiveness of cryoprecipitate in active bleeding and there is increasing interest in the use of fibrinogen concentrates. These are currently not licensed in the UK but have obvious advantages in light of their reduced infection risk and standardized fibrinogen concentration (Table 27.4). Case studies in trauma and RCTs in cardiac and urology surgery have reported positive outcomes following administration of fibrinogen concentrate (Table 27.5), but more evidence is needed as the studies are small and few are RCTs, and in no study has cryoprecipitate been directly compared to fibrinogen concentrate.
Table 27.4 Comparison of FFP and PCC, cryoprecipitate and fibrinogen concentrate.
Table 27.5 Studies evaluating the safety and efficacy of fibrinogen concentrate in trauma and major surgery.
Prothrombin complex concentrates (PCCs)
PCCs are plasma-derived coagulation factor concentrates that contain 3 or 4 vitamin K dependent factors at high concentration. PCCs contain 4 coagulation factors – II, VII, IX and X – as well as variable amounts of anticoagulants and heparin. PCCs are recommended for the treatment of serious or life-threatening bleeding related to oral anticoagulant therapy. Studies have shown that PCCs are safe and effective and normalize INR values rapidly when compared to FFP. Outcome data examining the effect of PCC on bleeding rates and mortality are not yet available. PCCs are currently licensed for treatment and perioperative prophylaxis of haemorrhage in patients with congenital and acquired deficiency of factors II, VII, IX or X, if purified specific coagulation factors are unavailable. PCCs are increasingly being considered as a substitute for FFP (Table 27.4 summarizes the differences between products) for use in coagulopathy associated with hepatic failure and traumatic haemorrhage. There is currently insufficient evidence to support these indications.
Massive blood transfusion (see also Chapter 26)
Strategies to manage massive blood transfusion have undergone major changes over the last few years, driven primarily by dismal outcomes observed in these patients using current treatment algorithms and evidence emerging from studies in battle casualties that higher volumes of FFP and platelets (approaching ratios of 1:1:1) lead to increased survival in massively transfused patients (Table 27.6). Some of these studies have demonstrated that early transfusion of FFP is key to favourable outcome rather than a high ratio per se. Importantly, there is no single FFP : RBC ratio that appears to be superior.
Table 27.6 Summary of observational trauma studies examining the effects of FFP : RBC ratios on outcomes.
These results have led to development of empirical early delivery of FFP and platelets in major haemorrhage protocols, with guidance of transfusion by coagulation testing later in the process. Major haemorrhage is, of course, not limited to trauma patients, but there is very little evidence to inform practice in other clinical settings. There are, however, clear differences between patient groups; many gastrointestinal haemorrhage patients are elderly, have limited cardiovascular reserve and may be susceptible to fluid overload. Massive haemorrhage protocols for trauma should not be applied to other clinical areas without significant consideration of patient comorbidities.
Major haemorrhage protocols
Between October 2006 and September 2010 delays in the provision of blood in UK hospitals led to 11 deaths and 83 incidents of harm being reposted to the National Patient Safety Agency. In light of this, a Rapid Response Report concerning the transfusion of blood in an emergency was produced, which recommended the adoption of major haemorrhage protocols in every hospital . Furthermore, the use of ‘drills’ to test local policies and ongoing education of all staff likely to be involved in massive haemorrhage protocols are advised. An example of a hospital protocol for major haemorrhage in trauma is given in Figure 27.5. Following data from the CRASH-2 study , tranexamic acid should be given to all adult trauma patients at risk of bleeding, so long as administration can be given within 3 hours of injury.
Fig 27.5 Example of a Massive Transfusion Protocol used in the setting of trauma. Courtesy of Dr R. Naidoo, Department of Haematology, John Radcliffe Hospital, Oxford.
The timely administration of blood components to patients with acute haemorrhage and in the critically ill is often life-saving; however, the clinical status of these patients also means that they are highly susceptible to organ dysfunction, and inappropriate transfusion with its associated risks may also have important adverse effects on clinical outcomes. Recent systematic reviews have identified important aspects of blood management that improve outcome as well as identifying gaps in knowledge that need to be addressed by future research. Restrictive transfusion practice is safe in critically ill patients without cardiovascular disease, and the safety of the approach in high risk groups is currently being evaluated in RCTs. Therapeutic adjuncts that reduce transfusion and improve outcomes are well defined and the wider application of these techniques will drive quality improvement. Coagulopathy associated with acute haemorrhage remains underresearched with no clear understanding of the underlying pathogenesis, accurate diagnostic tests or evidence-based treatments. The significant proportion of blood components utilized by these patients, their poor outcomes and significant utilization of healthcare resources are arguments for greater investment in this field.
1. Blood management focuses on improving patient outcomes in the setting of acute haemorrhage and acute anaemia in critically ill patients.
2. Restrictive use of allogeneic red cell transfusion is safe in patients without cardiovascular disease.
3. The use of techniques that enable autologous transfusion will reduce red cell transfusion and in the case of cell salvage will improve clinical outcomes and be cost effective.
4. Pharmacological blood conservation strategies effectively reduce transfusion exposure and in the setting of trauma improve survival.
5. Utilization of blood-conserving pharmacological strategies must consider the apparent increased risk of thromboembolic and other adverse clinical outcomes associated with greater efficacy in terms of reducing transfusion exposure, particularly in those at risk of cardiovascular complications.
6. Critically ill and acute surgical patients often develop coagulopathy. This is poorly defined and there are currently no validated sensitive and specific diagnostic tests that have been validated clinically or shown to improve clinical outcome.
7. The current evidence to support the prophylactic use of FFP and platelets in the critically ill is poor.
8. Massive transfusion protocols that place emphasis on communication, pre-emptive treatment and the use of higher platelet/FFP : RBC ratios appear to improve outcomes.
1. Wells AW, Llewelyn CA, Casbard A, Johnson AJ, Amin M, Ballard S, Buck J, Malfroy M, Murphy MF & Williamson LM. The EASTR Study: indications for transfusion and estimates of transfusion recipient numbers in hospitals supplied by the National Blood Service. Transfus Med 2009; 19: 315–327.
2. Corwin HL, Gettinger A, Pearl RG, Fink MP, Levy MM, Abraham E, MacIntyre NR, Shabot MM, Duh MS & Shapiro MJ. The CRIT Study: anemia and blood transfusion in the critically ill – current clinical practice in the United States. Crit Care Med 2004; 32: 39–52.
3. Bennett-Guerrero E, Zhao Y, O'Brien SM, Ferguson Jr TB, Peterson ED, Gammie JS & Song HK. Variation in use of blood transfusion in coronary artery bypass graft surgery. J Am Med Assoc 2010; 304: 1568–1575.
4. Carless PA, Henry DA, Carson JL, Hebert PP, McClelland B & Ker K. Transfusion thresholds and other strategies for guiding allogeneic red blood cell transfusion. Cochrane Database Syst Rev 2010; (10): CD002042.
5. Henry DA, Carless PA, Moxey AJ, O'Connell D, Forgie MA, Wells PS & Fergusson D. Pre-operative autologous donation for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev 2002; (2): CD003602.
6. Davies L, Brown TJ, Haynes S, Payne K, Elliott RA & McCollum C. Cost-effectiveness of cell salvage and alternative methods of minimising perioperative allogeneic blood transfusion: a systematic review and economic model. Health Technol Assess 2006; 10: iii–iv, ix–x, 1–210.
7. Carless PA, Henry DA, Moxey AJ, O'Connell D, Brown T & Fergusson DA. Cell salvage for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev 2010; (4): CD001888.
8. Zarychanski R, Turgeon AF, McIntyre L & Fergusson DA. Erythropoietin-receptor agonists in critically ill patients: a meta-analysis of randomized controlled trials. CMAJ 2007; 177: 725–734.
9. Alghamdi AA, Albanna MJ, Guru V & Brister SJ. Does the use of erythropoietin reduce the risk of exposure to allogeneic blood transfusion in cardiac surgery? A systematic review and meta-analysis. J Card Surg 2006; 21: 320–326.
10. Henry DA, Carless PA, Moxey AJ, O'Connell D, Stokes BJ, Fergusson DA & Ker K. Anti-fibrinolytic use for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev 2011; (3): CD001886.
11. Carless PA, Henry DA, Moxey AJ, O'Connell D, McClelland B, Henderson KM, Sly K, Laupacis A & Fergusson D. Desmopressin for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev 2004; (1): CD001884.
12. Levi M, Levy JH, Andersen HF & Truloff D. Safety of recombinant activated factor VII in randomized clinical trials. N Engl J Med 2010; 363: 1791–1800.
13. Ferraris VA, Brown JR, Despotis GJ, Hammon JW, Reece TB, Saha SP, Song HK, Clough ER, Shore-Lesserson LJ, Goodnough LT, Mazer CD, Shander A, Stafford-Smith M, Waters J, Baker RA, Dickinson TA, Fitzgerald DJ, Likosky DS & Shann KG. 2011 update to the society of thoracic surgeons and the society of cardiovascular anesthesiologists blood conservation clinical practice guidelines. Ann Thorac Surg 2011; 91: 944–982.
14. Corwin HL, Gettinger A, Fabian TC, May A, Pearl RG, Heard S, An R, Bowers PJ, Burton P, Klausner MA & Corwin MJ; EPO Critical Care Trials Group. Efficacy and safety of epoetin alfa in critically ill patients. N Engl J Med 2007; 357: 965–976.
15. CRASH-2 Trial collaborators, Shakur H, Roberts I, Bautista R et al. 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; 376(9734): 23–32.
16. SHOT (Serious Hazards of Transfusion), Annual Report; 2010. Available at: http://www.shotuk.org/wp-content/uploads/2011/07/SHOT-2010-Report.pdf.
17. Afshari A, Wikkelsø A, Brok J, Møller AM & Wetterslev J. Thrombelastography (TEG) or thromboelastometry (ROTEM) to monitor haemotherapy versus usual care in patients with massive transfusion. Cochrane Database Syst Rev 2011; (3): CD007871.
18. Arnold DM, Crowther MA, Cook RJ, Sigouin C, Heddle NM, Molnar L & Cook DJ. Utilization of platelet transfusions in the intensive care unit: indications, transfusion triggers, and platelet count responses. Transfusion 2006; 46: 1286–1291.
19. Stainsby D, MacLennan S, Thomas D, Isaac J & Hamilton PJ. British Committee for Standards in Haematology Guidelines on the management of massive blood loss. Br J Haem 2006; 135: 634–641.
20. Rossaint R, Bouillon B, Cerny V, Coats TJ, Duranteau J & Fernández-Mondéjar E. Task Force for Advanced Bleeding Care in Trauma. Management of bleeding for major trauma: an updated European guideline. Crit Care 2010; 14: R52.
21. Stanworth SJ, Walsh TS, Prescott RJ, Lee RJ, Watson DM & Wyncoll D; Intensive Care Study of Coagulopathy (ISOC) investigators. A national study of plasma use in critical care: clinical indications, dose and effect on prothrombin time. Crit Care 2011; 15: R108.
22. Yang L, Stanworth S, Hopewell S, Doree C & Murphy M. Is fresh frozen plasma clinically effective? An updated systematic review of randomised controlled trials. Transfusion 2012, 18 January. DOI: 10.1111/j.1537-2995.2011.03515.x. Epub ahead of print.
23. National Patient Safety Agency. The transfusion of blood and blood products in an emergency. Rapid Response Report, NPSA/2010/RRR017; 2010. Available at: http://www.nrls.npsa.nhs.uk/alerts/?entryid45=83659.
Blajchman MA, Glynn SA, Josephson CD & Kleinman SH; State-of-the-Science Symposium Transfusion Medicine Committee. Clinical trial opportunities in Transfusion Medicine: Proceedings of a National Heart, Lung, and Blood Institute State-of-the-Science Symposium. Transfus Med Rev 2010; 24: 259–285.
http://www.nba.gov.au/guidelines/review.html. Website of the state-wide blood management programme currently being implemented by the Department of Health of the Government of Western Australia.
Napolitano LM, Kurek S, Luchette FA, Corwin HL, Barie PS, Tisherman SA et al. American College of Critical Care Medicine of the Society of Critical Care Medicine; Eastern Association for the Surgery of Trauma Practice Management Workgroup. Clinical practice guideline: red blood cell transfusion in adult trauma and critical care. Crit Care Med 2009; 37: 3124–3157.