Patient blood management and perioperative anaemia

BJA Education, 17 (1): 28 34 (2017) doi: 10.1093/bjaed/mkw061 Matrix referenece 1I05, 2A05, 3I00 Patient blood management and perioperative anaemia S V Thakrar MBBS BSc(Hons) MRCP FRCA 1, *, B Clevenger MBBS BSc(Hons) MRCP FRCA 2, and S Mallett MBBS FRCA 3 1 StR 6 Anaesthesia, Clinical Research Fellow, Royal Free Hospital, London, 2 StR 6 Anaesthesia, Clinical Research Associate, Royal Free Hospital, London, and 3 Consultant Anaesthetist, Department of Anaesthesia, Royal Free Hospital, Pond Street, London, NW3 2QG, UK *To whom correspondence should be addressed. Tel: 07903556341; E-mail: s.thakrar1@nhs.net Key points Perioperative anaemia and allogenic blood transfusion are independent risk factors for poor postoperative outcomes: morbidity and mortality. Approximately one-third of patients are found to be anaemic on pre-assessment. Patient Blood Management (PBM) is a clinical concept with the goal of avoiding unnecessary blood transfusions to improve patient outcomes and safety. PBM involves early detection and treatment of preoperative anaemia, minimization of perioperative blood loss and improving tolerance to anaemia in the postoperative setting. Iron deficiency can be absolute or functional and requires iron supplementation to improve body reserves and utilization. PBM is a multimodal, multidisciplinary strategy requiring surgical and medical speciality input. anaemic at pre-assessment. Perioperative anaemia and allogenic blood transfusion are both, preventable surgical risks. Administration of blood in the perioperative setting is a risk factor which contributes to poor outcomes. 1 Patient Blood Management (PBM) is a clinical concept, which when implemented, has the primary goal of avoiding unnecessary blood transfusions and improving patient outcome and safety. This article summarizes PBM and the strategies involved in identifying and managing perioperative anaemia and blood transfusion. Patient blood management PBM involves a multidisciplinary approach to care, using evidence-based best practice in the management of perioperative anaemia and blood transfusion. It has been adopted by the World Health Organisation (WHO) and is supported by the National Blood Transfusion Committee in the UK. Recently published National Institute for Health and Care Excellence (NICE) (Table 1) guidelines for blood transfusion have incorporated many features of the PBM strategy including the consideration of alternatives to blood transfusion, active measures to reduce perioperative blood loss and the appropriate management of postoperative anaemia. 2 PBM focuses upon three pillars of care (Table 2): Perioperative anaemia is an independent risk factor for increased length of hospital and intensive care stay, postoperative complications, and increased mortality. 1 It is a strong predictor for perioperative blood transfusion requirements. In general, approximately one-third of patients are found to be 1. The optimization of red cell mass and erythropoiesis before operation. 2. The minimization of blood loss. 3. The management of postoperative anaemia. Implementation of PBM requires multidisciplinary involvement, and not infrequently organizational change, to facilitate these interventions. A National Comparative Audit of PBM Editorial decision: September 5, 2016; Accepted: September 18, 2016 VC The Author 2016. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved. For Permissions, please email: journals.permissions@oup.com 28

Table 1 Summary of recommendations NICE guidelines blood transfusion Alternatives to blood transfusion for patients having surgery RBCs Platelets Fresh frozen plasma (FFP) Prothrombin complex concentrate (PCC) Cryoprecipitate Patient safety Patient information within the UK 3 has provided benchmark standards for the implementation of BPM and the NICE transfusion guidelines provide clinicians with robust support for business cases when bidding for resource for service improvement. PBM identifies each patient s specific risks of anaemia, bleeding, and transfusion during the operative period. This allows patients to be better informed about blood transfusion and the alternatives, together with their risks. Detection and management of preoperative anaemia The WHO definition of anaemia states that it is a condition in which the number of red blood cells (RBCs) and their oxygen carrying capacity are unable to meet the body s physiological needs, 4 quantified by a haemoglobin concentration of less than 130 g dl 1 for men and less than 120 g dl 1 for non-pregnant women. The prevalence of anaemia before operation, has been found to be as high as 60%. 5 Anaemia has been shown to be an important modifiable risk factor for adverse outcomes after surgery. European Society of Anaesthesiology guidelines 6 recommend that haemoglobin be Offer oral iron in iron deficiency before and after surgery Offer TXA in surgery expected to have >500 ml blood loss Consider cell salvage with TXA if high volume blood loss expected Consider a threshold of 70 g litre 1 and a target of 70 90 g litre 1 after transfusion when using restrictive red blood cell transfusions Consider single-unit red blood cell transfusions for adults who do not have active bleeding In patients not bleeding or not having an invasive procedure or surgery: Offer prophylactic platelets with platelet count below 10 10 9 litre 1 and who do not have - Chronic bone marrow failure - Autoimmune thrombocytopenia - Heparin-induced thrombocytopenia - Thrombotic thrombocytopenic purpura Do not routinely transfuse more than a single dose of platelets Only consider FFP with clinically significant bleeding if coagulation tests are abnormal e.g. Prothrombin time ratio, Activated partial thromboplastin time ratio >1.5 Do not offer FFP to correct abnormalities in coagulation in patients who: - are not bleeding (unless having a procedure with a risk of significant bleeding) - Require reversal of vitamin K antagonist Offer immediate PCC for the emergency reversal of warfarin anticoagulation in: - severe bleeding or - head injury with suspected intracarebral haemorrhage Consider cryoprecipitate for patients with clinically significant bleeding and fibrinogen <1.5 g litre 1 Consider prophylactic cryoprecipitate for patients with fibrinogen level <1.0 g litre 1 who are having invasive procedures or surgery with a risk of bleeding Use 2 pools of cryoprecipitate and reassess the clinical condition Monitor for acute blood transfusion reactions Consider using electronic identification systems to improve safety and efficiency during the blood transfusion process Provide verbal and written information to patients who may have a transfusion explaining: - the reason for transfusion - the risks and benefits - the transfusion process - and specific transfusion needs - alternatives to transfusion - that they are no longer eligible to donate blood - that they are encouraged to ask questions measured 4 8 weeks before operation in all patients with an increased risk of bleeding to allow for adequate time for treatment of anaemia, if necessary, before surgery. Anaemia is conventionally categorized according to the mean cell volume (MCV). Anaemia can hence be defined as microcytic (MCV < 80 fl), macrocytic (MCV > 96 fl) or normocytic (MCV 80 96 fl). The primary causes of a microcytic anaemia are iron deficiency and congenital haemoglobinopathies. Rarer causes include sideroblastic anaemia, Vitamin B 6 deficiency, and lead poisoning. Macrocytic anaemia can be because of folate or vitamin B12 deficiency, medication, and alcoholism. Anaemia of chronic disease falls largely into the category of normocytic anaemia but may also be microcytic. Other causes of normocytic anaemia include aplastic and sickle cell anaemia, haemolysis, pregnancy, riboflavin, and pyridoxine deficiency. Anaemia can be directly related to the pathology requiring operative intervention e.g. gastrointestinal blood loss, uterine pathology, and colorectal cancers; or indirectly related, secondary to chronic disease, nutritional deficiency, haemoglobinopathy or renal impairment. BJA Education Volume 17, Number 1, 2017 29

Table 2 The three pillars of PBM Pillar 1 Detection and management of anaemia Pillar 2 Minimization of bleeding and blood loss Pillar 3 Manage anaemia and improve tolerance Preoperative Aim for assessment of anaemia Identify and manage bleeding risk (past 4 6 weeks before surgery medical and family history) Identify, evaluate and Review medications (antiplatelet, treat anaemia anticoagulation therapy) Treat absolute or functional iron Minimize iatrogenic blood loss deficiency with oral or i.v. iron Procedure planning and rehearsal Consider erythropoiesis stimulating agents if nutritional anaemia is ruled out/treated Refer for further evaluation as necessary Intraoperative Schedule surgery Meticulous haemostasis and with optimization surgical techniques of red cell mass Anaesthetic blood sparing techniques e.g. central neuraxial blockade Balanced physiology to aid optimal coagulation Patient positioning Goal directed management using point of care and VET Antifibrinolysis and cell salvage Postoperative Stimulate erythropoiesis Manage nutrition and correctable anaemia (e.g. avoid folate deficiency, iron restricted erythropoiesis) Iron deficiency can be because of depletion of the total iron stores or a chronic loss of blood. However, the metabolism of iron is also influenced by certain disease states including inflammation and malignancy. Raised iron stores and inflammation up-regulate the production of hepcidin, a hormone responsible for the inhibition of enteral iron absorption by degradation of the iron trans-membrane transporter ferroportin on duodenal enterocyte membranes. Hepcidin also inhibits the transport of stored iron from hepatocytes and macrophages into plasma in a similar manner. Thus the up-regulation of hepcidin can produce a functional iron deficiency, leading to what has traditionally been known as the anaemia of chronic disease. Haemoglobin is the principle transporter of oxygen and hence oxygen delivery is significantly reduced in anaemia. Compensatory mechanisms to maintain oxygen supply to the tissues include an increase in cardiac output and an increased oxygen extraction in the tissues facilitated by increased 2,3- diphosphoglycerate (2,3-DPG). This causes a rightward shift in the oxygen dissociation curve allowing oxygen to be released to the tissues more readily. The relative reduction in oxygen content is detected by tissue chemoreceptors leading to further compensation involving an increase in minute ventilation. Preoperative anaemia management Full blood count (FBC) should be assessed in all patients before listing for major elective surgery. 7 Ideally this should take place Monitor and manage bleeding Avoid secondary haemorrhage Maintain normothermia (unless specifically indicated) Autologous blood salvage Minimize iatrogenic blood sampling loss Haemostasis/anticoagulation management Compare estimated blood loss with patient specific tolerable blood loss Assess and optimize patient s physiologic reserve e.g. pulmonary and cardiac function Formulate patient-specific management plan using appropriate blood conservation modalities Optimize cardiac output Optimize oxygenation and ventilation Evidence based transfusion thresholds Maximize oxygen delivery Minimize oxygen consumption Avoid/treat infections promptly Evidence based transfusion thresholds 4 6 weeks before surgery. Patients who are found to be anaemic should have haematinics (those nutrients required for haematopoiesis) measured in order to diagnose and then appropriately treat the cause of the anaemia. Oral iron therapy is a low cost treatment for iron deficiency. However, patient compliance may be an issue as a result of the unwanted gastrointestinal side effects such as constipation. The bioavailability of oral iron is approximately 10 15% and up-regulation of hepcidin with its effects on ferroportin in duodenal enterocytes further reduces absorption. Thus, in a patient with inflammation or malignancy, absorption of enteral iron may be inadequate. Preoperative oral iron therapy has been shown to reduce blood transfusion requirements, however, therapy with good compliance for a period of 3 months or more may be required to adequately improve body iron stores. I.V. iron preparations have beenshowntobeeffectivein the treatment of iron deficiency anaemia with fewer side effects than oral preparations. 8 Newer preparations such as ferric carboxymaltose have improved safety profiles compared to older preparations, e.g. iron dextrans. A systematic review of i.v. iron therapy showed a decrease in allogenic blood transfusion, an increase in haemoglobin concentration and no significant increase in mortality and morbidity associated with infusion supporting this route of administration. 9 Local guidelines and pathways need to be in place to facilitate i.v. iron infusions. Pathways are required linking patients 30 BJA Education Volume 17, Number 1, 2017

referred to surgery, phlebotomy and blood testing, preoperative assessment services, and infusion facilities. These include haematology or renal day suites, who already provide outpatient infusions. Iron infusions must be given in appropriately staffed environments with facilities available to monitor patients and treat any complications, particularly anaphylaxis. An anaemia lead (usually an anaesthetist involved in preoperative assessment) is required to coordinate the pathway and multidisciplinary team. The pathway should have clear criteria for patients who are eligible, including locally agreed reference ranges for anaemia, operative severity, and bleeding risk. The final phase of the pathway should involve the rechecking of FBC to assess the response to the infusion, directing the patient to surgery or to further investigation and treatment. Erythropoiesis, the synthesis of RBCs, and consequently the production of haemoglobin are under the influence of erythropoietin (EPO), a hormone secreted by the kidney in response to cellular hypoxia. Exogenous, synthetic recombinant EPO can be used to stimulate RBC proliferation and it has been used successfully in patients with chronic kidney disease on dialysis. A systematic review of EPO used in patients undergoing cardiac or orthopaedic surgery has shown a reduction in the number requiring allogenic blood transfusion. The recent NICE guidance however, concluded that the benefit from a reduction in the numbers of patients transfused was offset by a potential increase in mortality and thrombotic complications. NICE recommended that EPO should not be used in surgical patients to reduce blood transfusion. Therefore a haematologist must be involved in any decision to use EPO for severe preoperative anaemia. Minimization of perioperative bleeding: management of coagulopathy and haemorrhage Preoperative assessment and management Preoperative risk stratification is important. A bleeding history should be obtained asking about past history of bleeding after surgery or trauma, childbirth, and menorrhagia in females, family history of bleeding diathesis and medication history particularly related to anticoagulants. The type of surgical intervention should also be considered with respect to the risk of bleeding and meticulous planning is required by the perioperative surgical team. Increasing numbers of patients continue anticoagulant and antiplatelet medications into the perioperative period. An individualized assessment of the thrombotic risk of stopping these drugs must be assessed against the risk of perioperative bleeding. This may require the substitution (bridging therapy) of long acting anticoagulants, such as warfarin, with shorter acting anticoagulants. Oral anticoagulant drugs including direct thrombin inhibitors (e.g. dabigatran) and direct factor Xa inhibitors (e.g. rivaroxaban and apixaban) are increasingly used in modern day practice as their pharmacokinetics are more predictable, removing the need for regular routine laboratory monitoring, unlike patients on warfarin. These newer drugs require preoperative discontinuation unless there is a low risk of bleeding. Many patients with coronary artery stents, particularly drug-eluting stents, need to continue anti-platelet agents into the perioperative period because of a high risk of in-stent thrombosis and myocardial infarction if medication is stopped. This must be balanced with the risk of bleeding when planning surgery. Planning can be assisted by the use of platelet function testing in order to stratify the risk of bleeding, whilst quantifying the level of platelet inhibition by antiplatelet agents. Intraoperative Intraoperative management of blood loss may be influenced by surgical technique, anaesthetic blood loss reduction strategies, and pharmacological management. Surgical technique is the key factor in determining perioperative blood loss. Minimally invasive surgery such as laparoscopic, robot assisted, and endovascular techniques are associated with a reduction in blood loss when compared with open more invasive approaches. Meticulous haemostasis is vital in preventing perioperative bleeding. The use of intraoperative topical haemostatic agents, containing collagens, fibrin or thrombin, may assist in the control of local bleeding. The use of regional anaesthesia is an anaesthetic blood sparing strategy. Central neuraxial blockade has been shown to significantly reduce perioperative blood loss. 10 Orthopaedic surgery remains the surgical speciality in which there is the most robust evidence for minimization of perioperative blood loss with the use of neuraxial blockade, and regional anaesthesia should be considered for all major joint surgery. Maintenance of a balanced physiology to aid optimal coagulation is essential: avoiding hypothermia, acidosis, and hypocalcaemia. Maintaining core body temperature above 35 C, ph > 7.2 and ionized calcium > 1 mmol litre 1 allows for optimum clot formation. 11 Manipulation of cardiovascular physiology to improve the operative field can reduce blood loss for example hypotensive anaesthesia in sinus surgery, or the lowering of central venous pressure in hepatic resection. Careful patient positioning will improve surgical operating conditions and reduce blood loss, by ensuring that venous drainage is maintained. Point of care testing Guided perioperative management of coagulopathy with the use of point of care testing has been shown to significantly reduce transfusion requirements. Conventional tests of coagulation have a long response time and can, in certain circumstances be misleading. 12 Haemostasis requires the presence of ample coagulation factors and platelets for sufficient thrombin generation and adequate fibrinogen levels (the substrate for fibrin production) to produce a stable clot. Viscoelastic testing (VET), including thromboelastography (TEG VR ) and rotational thromboelastometry (ROTEM VR ) measure changes in clot tensile strength over time, providing information on the dynamics of clot formation. VETs give a rapid description of the cell-based model of coagulation together with both cellular and humeral contributions to coagulation, see Figure 1. The administration of fresh frozen plasma, platelets, cryoprecipitate, factor concentrates and antifibrinolytic drugs can be guided by specific patterns of VET measurements. NICE guidelines currently only recommend the use of VET in cardiac surgery, but trials for guided therapy in trauma and obstetrics are ongoing. Cell salvage Where there may be blood loss greater than 1000 ml, the use of intraoperative cell salvage is advocated. Cell salvage is performed with the use of a double lumen suction device to collect blood. Blood is stored within a reservoir with added anticoagulants. Once enough blood is collected RBCs are washed, filtered, suspended in saline, and returned to the patient. The efficiency of cell salvage is improved with the use of antifibrinolytic drugs BJA Education Volume 17, Number 1, 2017 31

A C Distance B Distance Time Fig 1 VET. (A) Normal TEG trace: Grey annotation represents TEG, blue annotation represents ROTEM. R (Reaction time)/ct (Clotting time, K (K value)/cft (clot formation time), a (angle in degrees), MA (maximum amplitude)/mcf (maximum clot firmness) and Ly30 (lysis at 30 min as a ratio of MA)/CL (clot lysis index). (B) Prolonged R/ CT requiring coagulation factor replacement e.g. fresh frozen plasma. (C) Reduced MA/MCF requiring fibrinogen/platelet replacement. (D) Hyperfibrinolyis requiring antifibrinolytic agent e.g. TXA. and the use of tranexamic acid (TXA) when using cell salvage is recommended by NICE. 2 Antifibrinolytic drugs Antifibrinolytics drugs include TXA and e-aminocaproic acid. These inhibit fibrinolysis by their action at the active sites on plasminogen, inhibiting the activation of plasmin. TXA should be used prophylactically in major surgery where perioperative blood loss is predicted to be high. TXA has been shown to significantly reduce perioperative blood loss. The clinical randomization of an antifibrinolytic in significant haemorrhage (CRASH-2 trial) highlighted the benefit of early therapy with TXA in trauma induced bleeding. A significant decline in risk of death from haemorrhage with the early use of TXA in the trauma setting has been demonstrated. 13 This evidence has been extrapolated into other major intraoperative scenarios with the possibility of blood loss. Aprotinin is a potent antifibrinolytic agent of bovine origin that directly inhibits plasmin. It was used in cardiac procedures until it was withdrawn from clinical use after the blood conservation using antifibrinolytics (BART) randomized trial showed an increased risk of death with its use. 14 The results of that trial are disputed and subsequently aprotinin has been re-licenced for use in myocardial revascularization surgery. D Distance Time Time the incidence of anaemia. Reducing the frequency and volume of blood obtained for blood tests, particularly in the intensive care setting, can make a significant difference. Additional strategies include a reduction in the sample volume (e.g. using paediatric bottles), using continuous sampling lines with smaller dead space volumes (for arterial line samples) and evaluation of the requirement for each blood test according to clinical need. Blood loss into drains can be considerable. In most surgical specialties, the use of postoperative drains is being reduced, often as a component of enhanced recovery after surgery programmes. The use of drains has been shown to increase blood transfusion rates in orthopaedic surgery. Cell salvage can be utilized after operation, with the re-transfusion of blood from drains, particularly in major orthopaedic surgery. 15 The surgical stress response, inflammation and infection may precipitate functional iron deficiency in a similar manner to that of chronic disease. Before operation, both anaemic and non-anaemic patients (who may nonetheless be iron deficient) have been shown to respond to i.v. iron in the postoperative period. Consideration of replenishing iron stores should be made to reduce the incidence of postoperative anaemia and improve functional capacity for rehabilitation. Postoperative measures Efforts to reduce bleeding, avoid coagulopathy and minimize blood loss should be continued into the postoperative period. Simple interventions to decrease iatrogenic blood loss reduce Management of postoperative anaemia: improving tolerance with restrictive transfusion The body compensates for anaemia in an attempt to improve tissue oxygen delivery. Additional attempts to improve oxygen 32 BJA Education Volume 17, Number 1, 2017

deliverymaybemadebysupplementingthepatientwithoxygen and thereby increasing the inspired oxygen fraction. Maintenance of adequate organ perfusion with the use of vasopressors may also aid in the optimization of anaemia tolerance. Attempts should be made to decrease the relative oxygen demand by ensuring adequate analgesia and minimizing the chance of infection. Transfusion thresholds Large, randomized controlled trials have demonstrated that the use of restrictive transfusion of blood in the postoperative setting is safe and non-inferior to more liberal thresholds. The Transfusion Requirements in Critical Care study (TRICC) 16 comparing liberal vs restrictive blood transfusion demonstrated that restrictive blood transfusion, targeting a haemoglobin level of 70 90 g litre 1, is as effective as the use of liberal targets (Hb 100 120 g litre 1 ). In the Functional Outcomes in Cardiovascular Patients Undergoing Surgical Hip Fracture Repair study (FOCUS) liberal transfusion after major joint surgery did not improve mortality or functional capacity. 17 In critical care, patients with septic shock treated with a restrictive transfusion threshold of 70 g litre 1 had similar numbers of ischaemic events, severe adverse events and ventilatory support requirements compared to those with a threshold of 90 g litre 1. 18 Transfusion practice is now based upon much lower thresholds for transfusion when compared with historic traditional practice as evidence for the lower thresholds has strengthened. The European Transfusion Practice and Outcome Study reported a mean pre-transfusion Hb of 81 g litre 1. 19 NICE guidelines on blood transfusion recommend transfusion triggers of 70 or 80 g litre 1 in those with underlying cardiovascular disease. Single unit transfusion strategies Entrenched in clinical practice is the culture of 2-unit blood transfusions. In the face of increasing evidence to suggest little difference in mortality and morbidity associated with restrictive transfusion, single unit blood transfusion is recommended in stable patients without active bleeding. Each unit of blood transfused is thought of as an independent clinical decision and subsequent units of blood are based on a post-transfusion FBC. A single unit blood transfusion policy reduces transfusion rates and improves the economics of blood transfusion. Practical implementation of PBM The benefit of establishing PBM in perioperative pathways is clear. From a clinical perspective, the optimization of patient performance through the treatment and prevention of anaemia reduces the exposure of patients to the risks of transfusion. A reduction in transfusion rates and their associated costs saves scarce resources, which may be redeployed. All hospitals have a Hospital Transfusion Committee (HTC), and it is now recommended that a multidisciplinary PBM team be established alongside the HTC. Engagement of hospital management is essential to overcome barriers to PBM implementation. Benchmarking of performance including transfusion rates can help to identify particular areas of need, whilst preparation of business cases for PBM initiatives require management backing. Data collection and audit of progress is essential to the continued success of such initiatives. Parameters which should be monitored include: use of blood components against local and national guidelines; proportions of blood components transfused with pre-transfusion blood results and clinical indication documentation; proportion of preoperative patients screened for anaemia before surgery where blood transfusion is most likely. The adaptation of surgical pathways to encourage timely pre-assessment can facilitate anaemia treatment. Greater staff education and engagement can subsequently improve the education and engagement of patients with respect to the risks and benefits of blood transfusion and improve rates of informed consent for blood transfusion. Conclusion PBM is a multimodal, multidisciplinary approach to reducing the risks posed by perioperative anaemia and allogenic blood transfusion. Many of the strategies underpinning PBM are highlighted in recently published national guidelines and the implementation of PBM has been shown to have both patient and economic benefit. MCQs The associated MCQs (to support CME/CPD activity) can be accessed at https://access.oxfordjournals.org by subscribers to BJA Education. References 1. Musallam KM, Tamim HM, Richards T et al. Preoperative anaemia and postoperative outcomes in non-cardiac surgery: a retrospective cohort study. The Lancet 2011; 378: 1396 407 2. NICE guidelines: blood transfusion. 2015, https://www.nice. org.uk/guidance/ng2 (accessed 23 November 2015) 3. National comparative audit of blood transfusion, http:// hospital.blood.co.uk/audits/national-comparative-audit/ (accessed 23 November 2015) 4. WHO: haemoglobin concentration for the diagnosis of anameia and assessment of severity, http://www.who.int/ vmnis/indicators/haemoglobin/en/ (accessed 23 November 2015) 5. Shander A, Knight K, Thurer R, Adamson J, Spence R. Prevalence and outcomes of anemia insurgery:a systematic review of the literature. Am J Med 2004; 116: 58S 69S 6. Kozek-Langenecker SA, Afshari A, Albaladejo P et al. Management of severe perioperative bleeding: guidelines from the European Society of Anaesthesiology. Eur J Anaesthesiol 2013; 30: 270 382 7. Klein AA, Arnold P, Bingham RM et al. AAGBI guidelines: the use of blood components and their alternatives 2016. Anaesthesia 2016; 71: 829 42 8. Tolkien Z, Stecher L, Mander AP, Pereira DI, Powell JJ. Ferrous sulfate supplementation causes significant gastrointestinal side-effects in adults: a systematic review and meta-analysis. PLoS One 2015; 10: e0117383 9. Litton E, Xiao J, Ho KM. Safety and efficacy of intravenous iron therapy in reducing requirement for allogeneic blood transfusion: systematic review and meta-analysis of randomised clinical trials. BMJ 2013; 347: f4822 10. Richman JM, Rowlingson AJ, Maine DN, Courpas GE, Weller JF, Wu CL. Does neuraxial anesthesia reduce intraoperative blood loss? A meta-analysis. JClinAnesth2006; 18: 427 35 BJA Education Volume 17, Number 1, 2017 33

11. Lier H, Krep H, Schroeder S, Stuber F. Preconditions of hemostasis in trauma: a review. The influence of acidosis, hypocalcemia, anemia, and hypothermia on functional hemostasis in trauma. J Trauma 2008; 65: 951 60 12. Mallett SV, Chowdary P, Burroughs AK. Clinical utility of viscoelastic tests of coagulation in patients with liver disease. Liver Int 2013; 33: 961 74 13. Collaborators CT. 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: 23 32 14. Fergusson DA, Hébert PC, Mazer CD et al. For the BART Investigators. N Engl J Med 2008; 358: 2319 31 15. Ashworth A, Klein AA. Cell salvage as part of a blood conservation strategy in anaesthesia. Br J Anaesth 2010; 105: 401 16 16. Hébert PC, Wells G, Blajchman M et al. A Multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med; 340: 409 17 17. Carson JL, Terrin ML, Noveck H et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med 2011; 365: 2453 62 18. Holst LB, Haase N, Wetterslev J et al. Lower versus higher hemoglobin threshold for transfusion in septic shock. N Engl J Med 2014; 371: 1381 91 19. Meier J, Filipescu D, Kozek-Langenecker S et al. Intraoperative transfusion practices in Europe. Br J Anaesth 2016; 116: 255 61 34 BJA Education Volume 17, Number 1, 2017