Blood is altered from the moment of collection and the ‘lesions’ of collection – anticoagulation, separation, cooling, preservation and storage – compound and progressively increase until the date of expiry.³ The extent of these changes is determined by collection technique, the specific blood component, the preservative medium, the container, storage time and storage conditions. The threshold storage time for blood components has generally been arbitrarily determined by in vitro studies and assessment of in vivo survival. In the case of red cell concentrates, greater than 75% of transfused cells should survive post transfusion.

Storage results in quantitative and/or qualitative deficiencies in blood components, which may reduce the immediate efficacy of a transfusion. In parallel with these storage changes is an accumulation of degenerate material (e.g. microaggregates and procoagulant material), release of vasoactive agents, cytokine generation and haemolysis.⁴ Many of the changes occurring during storage are related to the presence of leukocytes and can be minimised by pre-storage leukoreduction.⁵ Red cells undergo a change from their biconcave disc shape to spiky spherocytes (echinocytes) and in so doing lose their flexibility. There are also changes in the red cell membrane resulting in an increased tendency to adhere to endothelial cell surfaces in the microcirculation, especially if there is activation of endothelial cells, for example in the presence of the systemic inflammatory response (e.g. with shock or sepsis).⁶ There is evidence that the immediate post-transfusion function of stored red cells and haemoglobin in delivering oxygen to the microcirculation and unloading is questionable, and several hours are required for red cell oxygen carriage and delivery to return to normal.^7,⁸ It is important to differentiate between the storage lesion being responsible for failure to achieve clinical/laboratory end-points due to reduced survival and/or qualitative defects in cellular function and the ‘toxic’ effects of blood storage (Figure 88.2).

Figure 88.2 The storage lesions. ARDS, adult respiratory distress syndrome; GIT, gastrointestinal tract; MOF, multiorgan failure; RES, reticuloendothelial system; TRALI, transfusion-related acute lung injury.

The use of blood filters has been an acknowledgement of the existence of the blood storage lesion and its possible clinical significance. The 170 μm blood-giving filters were first introduced into transfusion medicine to stop the occlusion of blood-giving sets. Ironically, there was little concern that the fibrin clots may harm the patient but fortunately the lung is one of nature’s remarkable filters. Adult respiratory distress syndrome (ARDS) and the Vietnam War increased interest in unfiltered microaggregates accumulating during storage. Both logic and animal data suggested their implication in ARDS and that microfilters to remove microparticles 20–40 μm in size may be protective. This proved difficult to confirm but microaggregate filters do not adequately address the problem of the storage lesion and its clinical significance. Using pre-storage leukodepletion filters, and preventing the development of the storage lesion in both blood and platelets from its inception, is more logical and scientific. Universal pre-storage leukoreduction is now standard practice in many countries, although it was primarily introduced as a precautionary measure against the possible transmission of variant Creutzfeldt–Jakob disease (vCJD) and not to address the numerous other indications for the removal of leukocytes.⁹

The clinical significance of blood storage lesions is still controversial. Further studies are needed to assess their relevance in conditions such as ARDS, multiorgan failure (MOF), vasoactive reactions and alterations in laboratory parameters.^7,^10,¹¹ It is assumed that blood components have been appropriately collected, processed, stored, transported and transfused, but despite much greater attention to standard operating procedures and regulation generally, the quality of the final product cannot be guaranteed.¹² The ‘assumed’ quality of labile cellular blood products is based on research data and monitoring of standard operating procedures. There is rarely detailed individual product assessment prior to transfusion.⁸ It is accepted that the adverse effects of storage increase with time and an arbitrary ‘cut off’ is mandated on the basis of research studies.

In relation to the possible clinical significance of the storage lesion, the following should be considered:

• Quantitative and qualitative deficiency of blood component

• Failure to achieve anticipated end-points due to reduced quantity and/or quality of the blood product

• Exposure to excessive numbers of donors in achieving efficacy

• Physical characteristics

• Hypothermia

• Chemical characteristics

• Citrate toxicity

• Acid–base imbalance

• Glucose

• Contamination

• Bacterial resulting in endotoxaemia or septicaemia

• Plasticisers

• Accumulation of ‘toxic’ or degenerate products

• Role of the storage lesion in transfusion-related immunomodulation

• Role of cytokines

• Role of reticuloendothelial system blockade

• Accentuation of free radical pathophysiology due to free iron

• Effects of transfusion on laboratory parameters (e.g. elevations in bilirubin, neutrophils, serum iron and lactic dehydrogenase) which may lead to incorrect interpretation

• Large volume transfusions (proportional to storage age) as a risk factor for MOF and ARDS

• Early hyperkalaemia, late hypokalaemia

• Activation and consumption of the haemostatic factors with possible contribution to disseminated intravascular coagulation (DIC) and venous thromboembolism

• Non haemolytic, non-febrile transfusion reactions

• Hypotension and circulatory instability due to vasoactive substance (kinins, histamine)

THE ROLE OF LEUKOCYTES AS A ‘CONTAMINANT’ IN LABILE STORED BLOOD AND THE ROLE OF PRE-STORAGE LEUKODEPLETION

There are several difficulties assessing potential adverse effects of leukocytes as they may be responsible for a wide range of blood component quality and safety issues.⁹ The presence of leukocytes has been shown to be responsible for specific adverse outcomes in some patients (e.g. non-haemolytic febrile transfusion reaction, platelet refractoriness and transfusion-associated graft-versus-host disease, TAGVHD), but this is the minority. In the larger context the overall available evidence, in the absence of adequate large randomised clinical trials, suggests that universal pre-storage leukodepletion may reduce transfusion-related morbidity and mortality as well as generating cost savings.¹³ Leukodepletion of red cell and platelet concentrates minimises the clinical consequences of the immunomodulatory effects of allogeneic transfusion. Hence it may decrease the incidence of recurrence of some cancers, of postoperative infections, of blood stream infections and reduce ICU and hospital length of stay. In many patients transfusion-related acute lung injury (TRALI) is a multifactorial disorder and in ‘at risk’ patients non-leukodepleted blood may be a risk factor. Patients in whom there is activation of the systemic inflammatory response syndrome (SIRS) are at risk of developing the multiorgan failure syndrome.^5,¹⁴ Patients at particular risk include those with trauma, burns, critical bleeding, shock, sepsis and those undergoing cardiopulmonary bypass.¹⁵^–¹⁹ The quality and function of pre-storage leukodepleted red cell concentrates is better maintained on storage, ensuring better post-transfusion efficacy and survival.^6,²⁰

CLINICAL GUIDELINES FOR BLOOD COMPONENT THERAPY

The following is a brief summary of the clinical guidelines for the use of commonly used blood components. The use of specific concentrates or recombinant products is beyond the scope of this book.

RED CELL TRANSFUSIONS

What constitutes appropriate use of red cell transfusions in acute medicine is contentious because of the difficulties in identifying the benefits of red cell transfusion in many circumstances.²¹ Pursuit of the lowest safe haematocrit continues to receive considerable attention, but pushing any aspect of a system to its limits risks ‘sailing too close to the wind’ which may be appropriate in some situations, but potentially hazardous in others.^22,²³

In an otherwise stable patient, the transfusion of red cell concentrates is likely to be inappropriate when the hemoglobin level is > 100 g/l. On the other hand, their use may be appropriate when haemoglobin is in the range 70–100 g/l if there are other defects in the oxygen transport system, such as cardiorespiratory dysfunction. The decision to transfuse should be supported by the need to relieve clinical signs and symptoms of impaired oxygen transport and prevent morbidity and mortality. The transfusion of red cell concentrates is likely to be appropriate when hemoglobin is < 70 g/l and the anaemia is not reversible with specific therapy in the short term, but lower levels may be acceptable in patients who are asymptomatic, especially in the younger age group.

PLATELET TRANSFUSIONS

Platelet transfusion therapy may benefit patients with platelet deficiency or dysfunction. The following are the indications for platelet transfusions:²⁴

PROPHYLAXIS

• Bone marrow failure when the platelet count is < 10 × 10⁹/l without associated risk factors for bleeding or < 20 × 10⁹/l in the presence of additional risk factors

• To maintain the platelet count at > 50 × 10⁹/l in patients undergoing surgery or invasive procedures

• In qualitative platelet function disorders, depending on clinical features and setting (platelet count is not a reliable indicator for transfusion)

BLEEDING PATIENT

• In any haemorrhaging patient in whom thrombocytopenia secondary to marrow failure is considered a contributory factor

• When the platelet count is < 50 × 10⁹/l in the context of massive haemorrhage/transfusion and < 100 × 10⁹/l in the presence of diffuse microvascular bleeding

The transfusion of platelet concentrates is not generally considered appropriate:

• when thrombocytopenia is due to immune-mediated destruction

• in thrombotic thrombocytopenic purpura and haemolytic uraemic syndrome

• in uncomplicated cardiac bypass surgery

FRESH FROZEN PLASMA

There are few specific indications for fresh frozen plasma, but its use may be appropriate:²⁵

• for replacement of single factor deficiencies where a specific or combined factor concentrate is not available

• for treatment of the multiple coagulation deficiencies associated with acute DIC

• for treatment of inherited deficiencies of coagulation inhibitors in patients undergoing high-risk procedures where a specific factor concentrate is unavailable

• in the presence of bleeding and abnormal coagulation parameters following massive transfusion or cardiac bypass surgery or in patients with liver disease

• for immediate reversal of warfarin-induced anticoagulation in the presence of potentially life-threatening bleeding when prothrombin concentrates (PCC) are not available ²⁶^–²⁸ (see Chapter 91)

The use of fresh frozen plasma is generally not considered appropriate in cases of:

• hypovolaemia

• plasma exchange procedures unless post exchange invasive procedures are planned

• treatment of immunodeficiency states

CRYOPRECIPITATE

Cryoprecipitate is prepared from freshly collected plasma and contains factor VIII, fibrinogen, von Willebrand factor, factor XIII and fibronectin. Cryoprecipitate was originally developed for treatment of haemophilia due to factor VIII deficiency, but is now prescribed to treat congenital or acquired hypofibrinogenaemia, usually in the context of liver disease, trauma, DIC, hyperfibrinolysis or massive transfusion.

IMMUNOGLOBULIN

Normal human immunoglobulin is available in intramuscular and intravenous forms for the treatment or prevention of infection in patients with proven hypogammaglobulinaemia.²⁹ Intravenous immunoglobulin G has been recommended as adjunctive therapy in patients with fulminant sepsis syndrome, especially those with toxic shock syndrome caused by group A streptococci.³⁰

Intravenous immunoglobulin therapy also has a role in therapy of some autoimmune disorders, such as idiopathic thrombocytopenic purpura, autoimmune polyneuropathy and others (see Chapter 90).

FACTOR CONCENTRATES

There is an increasing number of plasma protein concentrates available for clinical use. Some are prepared from donor plasma and some by recombinant technology. Factor VIII and factor IX concentrates have an established role in the management of haemophilia, but others are in the process of establishing their clinical efficacy and indications. Antithrombin III (ATIII) concentrates are available for thrombophilia due to ATIII deficiency and are increasingly recommended in other disorders where ATIII may be depleted (e.g. DIC, MOF).³¹ Recombinant human activated protein C has antithrombotic, anti-inflammatory and profibrinolytic properties but its role in the treatment of patients with severe sepsis remains controversial.³²

Of recent interest is the use of the recombinant activated haemostatic proteins and inhibitors. Recombinant activated factor VII (rFVIIa) was developed for the management of haemophilic patients with coagulation factor inhibitors. rFVIIa is now being widely used ‘off label’ for controlling haemorrhage in the non-haemophilic setting and there is considerable controversy as to its benefits and risks.³³ There are an increasing number of case reports and series reporting on the use of rFVIIa in critical life-threatening haemorrhage. Patients with uncontrolled critical bleeding and coagulopathy, despite large transfusions and surgical intervention, have significant mortality rates and rFVIIa in these clinical situations is used as salvage therapy. rFVIIa is now being studied in controlled trials where there is critical bleeding in various clinical settings. rFVIIa initiates the extrinsic coagulation pathway when complexed to tissue factors at sites of injury and on activated platelet surfaces. It potentially has a role in a wide range of haemostatic disorders (e.g. massive blood transfusion, liver disease, uraemia, severe thrombocytopenia and platelet disorders).³⁴

POTENTIAL ADVERSE EFFECTS OF ALLOGENEIC TRANSFUSION

The pathophysiology of blood transfusion reactions can broadly be divided into five categories (Figure 88.3):

• Reactions may occur due to immunological differences between the donor and recipient, resulting in varying degrees of blood component incompatibility. In general, in order for a reaction to occur, the recipient needs to have been previously immunised to a cellular or plasma antigen.³⁵

• A wide range of infectious agents may be transmitted by homologous blood component therapy.

• Alterations in blood products due to preservation and storage may result in quantitative and/or qualitative deficiencies in the blood components which will reduce transfusion efficacy and expose the patient to potentially adverse consequences from storage accumulants in the component.

• Technical or clerical errors that may result in adverse events in all the above categories.

• Psychological – informed consent issues and patient anxiety compatibility.

Figure 88.3 Possible mechanisms of adverse effects to blood transfusion. TAGVHD, transfusion-associated graft-versus-host disease.

In terms of causation of an adverse clinical event, the possible role of blood transfusion can be classified into three categories on the basis of probability (Figure 88.4):

• Definite – unifactorial – transfusion caused: The well-understood and reported hazards of transfusion (i.e. immunological, technical, infectious) are generally unifactorial with a 1:1 causal relationship between the blood component transfused (usually a specific individual unit) and the adverse consequence for the patient. ABO blood group incompatibility, transfusion-related infection transmission, TAGVHD and TRALI due to donor leukoagglutinins are examples in this category.

• Probable – oligofactorial – transfusion related: Some adverse consequences of transfusion result from interaction with other insults, pathophysiology or host factors, but the contribution of the transfusion can usually be specifically identified. Fever, allergic reactions, hypotensive reactions, pulmonary oedema, some cases of TRALI, hyperbilirubinaemia and cytomegalovirus (CMV) transmission are examples in this category.

• Possible – multifactorial – transfusion associated: Transfusion may be a contributor to a complication or poor clinical outcome. In these circumstances it may be difficult to implicate transfusion directly in an individual case, nor is it necessarily the major factor. Transfusion-induced immunomodulation (TRIM) and the clinical consequences of storage lesions fall into this category. The role of transfusion being associated with adverse consequences has been established in observational studies, but causation cannot be confidently concluded. Accumulating evidence, in the absence of randomised clinical trials, supports causation, thus demanding a more precautionary approach to allogeneic blood transfusion, examination of alternatives to transfusion and implementation of methods to improve the quality of transfused blood components (especially red cell and platelet concentrates). Universal pre-storage leukodepletion is currently the most important and effective strategy for minimising the clinical impact of TRIM and the blood storage lesions.

Figure 88.4 Algorithm for analysis for the possibility of a transfusion reaction. ARDS, adult respiratory distress syndrome; CMV, cytomegalovirus; MOF, multiorgan failure; TAGVHD, transfusion-associated graft-versus-host disease; TRIM, transfusion-induced immunomodulation; TRALI, transfusion-related acute lung injury.

Adverse reactions to blood component therapy may present in a wide range of ‘guises’ and transfusion must always be considered in the differential diagnosis of any unexpected clinical presentation or indeed as a contributory factor to a clinical picture that may have several contributory factors (Figure 88.5).

Figure 88.5 Adverse reactions to blood component therapy may just be one of the contributory factors in the differential diagnosis of any unexpected clinical presentation.

PYREXIA

Mild febrile reactions are not usually a matter of concern, but rigors and temperatures above 38°C should not be ignored. The majority of febrile reactions are now considered to be an immunological reaction against one or more of the transfused cellular or plasma components, usually leukocytes.

TRANSFUSION-RELATED INFECTIONS

OTHER TRANSFUSION-TRANSMISSIBLE INFECTIONS

There is an increasing range of agents that have the potential to be transmitted by blood transfusion, the most recent one of concern being vCJD. The reader is referred to reviews for further information.^36,³⁷

BACTERIAL CONTAMINATION

Bacterial contamination of stored blood has always been recognised as a potential cause of fulminant endotoxic shock. Although a rare complication, continuing reports of its occurrence, especially in relation to platelet concentrate transfusion, have increased awareness.^38,³⁹ The clinical features of transfusion-related bacterial sepsis in the non-anaesthetised patient include rigors, fever, tachycardia and vascular collapse, with prominent nausea, vomiting and diarrhoea. Anaesthetised patients may have a delayed onset of symptoms (fever, tachycardia, hypotension and cyanosis), followed by DIC, renal failure and sometimes ARDS. The requirement to store platelet concentrates at room temperature increases the risks of bacterial contamination. Pre-release bacterial testing is being more widely advocated as a measure to minimise the problem.

HAEMOLYTIC TRANSFUSION REACTIONS

Most severe acute haemolytic transfusion reactions are due to ABO incompatibility, have an identifiable cause and are avoidable, in contrast to delayed haemolytic reactions which are immune in nature and usually cannot be prevented.

INITIAL SYMPTOMS AND SIGNS

The classic symptoms and signs of an acute haemolytic transfusion reaction, typical of ABO incompatibility, include apprehension, flushing, pain (e.g. at infusion site, headache, chest, lumbosacral and abdominal), nausea, vomiting, rigors, hypotension and circulatory collapse.

HAEMOSTATIC FAILURE

Haemorrhagic diathesis due to DIC may be a feature, resulting in severe generalised haemostatic failure, with haemorrhage and oozing from multiple sites.⁴⁰ As the responsible transfusion is likely to have been administered for haemorrhage, increasing severity of local bleeding may be the first clue to an incompatible transfusion, especially if the patient is unconscious or anaesthetised in the operating room.

OLIGURIA AND RENAL IMPAIRMENT

Renal impairment may complicate a haemolytic transfusion reaction and its prevention or the appropriate management of established renal failure is important. If circulating volume and urinary output are rapidly restored, established renal failure is unlikely to occur. Death from acute renal failure directly caused by an incompatible blood transfusion is preventable, and there are usually other poor prognostic factors.

ANAEMIA AND JAUNDICE

A severe haemolytic transfusion reaction may be suspected from the development of jaundice or anaemia.

ALLERGIC AND ANAPHYLACTOID REACTIONS

Non-cellular blood (plasma and plasma derivatives) components are rarely considered to be a major cause for adverse reactions to transfusion therapy. However, the complexity of plasma and its various components and the effects from component processing results in a broader spectrum of potential adverse effects than frequently recognised. The antigenic heterogeneity of plasma proteins and the presence of antibodies make the non-cellular components of blood responsible for a range of adverse effects, many of which remain poorly understood and commonly pass unrecognised in clinical practice.⁴¹

There has been debate over the years as to the classification of allergic reactions to blood components. Clinical severity may range from minor urticarial reactions or flushing through to fulminant cardiorespiratory collapse and death. Some of these reactions are probably true anaphylaxis, but in others, mechanisms have been less clear and the term anaphylactoid has been used. To avoid implying the mechanism of the reaction, the term immediate generalised reaction has been used.

The clinical syndromes of immediate reactions have been classified as follows:

• Grade I

• skin manifestations

• Grade II

• mild to moderate hypotension

• gastrointestinal disturbances (nausea)

• respiratory distress

• Grade III

• severe hypotension, shock

• bronchospasm

• Grade IV

• cardiac and/or respiratory arrest

PLASMA AND PLASMA COMPONENTS MAY CAUSE ADVERSE EFFECTS BY SEVERAL PATHOPHYSIOLOGICAL MECHANISMS

Immunological reactions to normal components of plasma may occur in two ways:

• Plasma proteins being antigenic to the recipient; they may contain epitopes on their molecules different from those on the recipient’s functionally identical plasma proteins (e.g. anti-IgA antibodies)

• Antibodies in the donor plasma reacting with cellular components of the recipient’s blood cells or plasma proteins

Physicochemical characteristics and contaminants of donor plasma, such as temperature, chemical additives, medications and micro-organisms may be responsible for recipient reactions.

The preparation techniques and storage conditions of blood and blood products may potentially cause adverse reaction through:

• accumulation of metabolites or cellular release products

• plasma activation, i.e. activation of some of the proteolytic systems. Importantly, the complement and kinin/kininogen systems may generate vasoactive substances and anaphylatoxins which may be responsible for reactions. Some apparently allergic reactions to blood products may be due to vasoactive substances in the infusion. Subjective sensations may be missed in an unconscious patient. Hypotension occurring during rapid infusion of a hypovolaemic patient is likely to be interpreted as further volume loss, particularly with some plasma protein fractions that have been reported as consistently producing a transient fall in blood pressure, and a situation fraught with risk of overload can be produced

• histamine generation: histamine levels increase in some stored blood components and levels may be correlated with non-febrile, non-haemolytic transfusion reactions. Histamine release may be stimulated in the patient by plasma components, synthetic colloids and various medications

• generation of cytokines during storage may be responsible for non-haemolytic transfusion reactions

• chemical additives: there are various chemical additives (ethylene oxide, formaldehyde, drugs, latex) which may be responsible for immunological or non-immunological recipient reactions.

TRANSFUSION-ASSOCIATED GRAFT-VERSUS-HOST DISEASE (TAGVHD)

Graft-versus-host disease, classically observed in relationship to allogeneic bone marrow transplantation, may occur following a blood transfusion due to the infusion of immunocompetent lymphocytes precipitating an immunological reaction against the host tissues of the recipient.⁴² It is most commonly observed in immunocompromised patients, but may also be seen in recipients of directed blood donation from first-degree relatives, and occasionally where donor and recipient are not related, due to homozygosity for HLA haplotypes for which the recipient is heterozygous. The syndrome usually occurs 3–30 days post allogeneic transfusion with fever, liver function test abnormalities, profuse watery diarrhoea, erythematous skin rash and progressive pancytopenia.

TRANSFUSION-RELATED ACUTE LUNG INJURY (TRALI)

TRALI is receiving a great deal of attention as a potentially serious complication of blood transfusion.⁴³^–⁴⁶ In the classic plasma neutrophil antibody-mediated form of the disease symptoms usually arise within hours of a blood transfusion.⁴⁷ In contrast to most patients with ARDS, recovery usually occurs within 48 hours. The underlying pathophysiology of ‘classic’ TRALI is due to the presence of leukoagglutinins in donor plasma. When complement is activated, C5a promotes neutrophil aggregation and sequestration in the microcirculation of the lung, causing endothelial damage leading to an interstitial oedema and acute respiratory failure. This classic form of transfusion-related acute lung injury has been recognised for over five decades, but was thought to be a rare complication of allogeneic plasma transfusion. As leukoagglutinins occur typically in plasma from multiparous females there is a move to using male donor only fresh frozen plasma.

It is now accepted that there has been underrecognition and underdiagnosis of TRALI, partly due to a lack of clinical awareness, but also to a broader understanding of mechanisms by which blood transfusion may cause or be a contributory factor to lung injury.^48,⁴⁹ The term TRALI is now being expanded to include cases in which transfusion is identified as an independent risk factor predisposing patients to lung injury.^50,⁵¹ The reader is referred to recent reviews addressing the expanding area of concern in which transfusion, per se, is being identified as an independent risk factor for poor clinical poor outcomes, of which TRALI is only one.

TRANSFUSION-RELATED IMMUNOMODULATION (TRIM)

There is evidence that allogeneic blood transfusion is immunosuppressive in the recipient that has implications in relation to resistance to infection and the likelihood of cancer recurrence.⁵²^–⁵⁴ Donor leukocytes probably play the most important role in this immunomodulation. TRIM, along with the clinical effects of the blood storage lesion, is probably the mechanism by which allogeneic blood transfusion is responsible for poorer clinical outcomes discussed in the next section.

ALLOGENEIC TRANSFUSION AS AN INDEPENDENT RISK FACTOR FOR POORER CLINICAL OUTCOMES

Over the last decade experimental and clinical studies have identified blood transfusion as an independent risk factor for morbidity and mortality as well as increased admission rates to ICUs, increased length of hospital stay and additional costs.^18,^19,⁵⁵^–⁵⁷ Blood transfusion being implicated as part of the problem rather than optimal therapy has been a surprise to many clinicians, health administrators and patients, as it has always been assumed that blood transfusion can only be of benefit to the bleeding or anaemic patient. This long-standing belief is now being effectively challenged and currently the debate on the management of the acutely haemorrhaging or anaemic patient has moved significantly from a paradigm with a focus on transfusion to one in which the urgent control of critical bleeding is the paramount priority, with avoidance and/or minimisation of blood transfusion. There is thus increasing evidence that transfusion-related immunomodulation (TRIM) and the transfusion effects of storage lesions are responsible for poorer clinical outcomes in a range of clinical settings.^18,^19,^55,⁵⁸^–⁶³

However, it is not possible to categorically conclude causation and that pre-storage leukoreduction will eliminate or minimise the risk. While the case for causation is strengthened, it is important to be aware that there is minimal evidence for the efficacy or appropriateness of many transfusions, and there is evidence that restrictive red cell transfusion policies do not jeopardise clinical outcomes; indeed, the reverse may be the case.^64,⁶⁵ Thus, until these issues are resolved, a precautionary approach should be adopted, with the modus operandi being to avoid or minimise allogeneic transfusion and use appropriate patient blood conservation techniques.⁶⁶ An additional association with transfusion is the higher incidence of venous thromboembolism if patients are transfused.⁶⁷

CRITICAL HAEMORRHAGE AND MASSIVE BLOOD TRANSFUSION

There is a reappraisal of clinical practice guidelines for the use of blood components in the bleeding patient.^21,⁶⁸ The guidelines are no longer for the management of massive blood transfusion, but rather for the management of critical bleeding and avoiding getting into the massive transfusion coagulopathy quagmire in which the patient can spiral down into the ‘triad of death’ – coagulopathy, acidosis and hypothermia.

This reappraisal of management of patients with acute haemorrhage is resulting in the challenging of long-standing dogmas. There is greater tolerance of hypotension until haemorrhage is controlled, and lower haemoglobin levels are tolerated with closer attention to the clinical context of the anaemia and its impact on systemic and local oxygen delivery, especially if there are any compromises in cardiorespiratory function. More attention and research is being directed toward the role of clear fluids and the importance of plasma viscosity, colloid oncotic pressure and functional capillary density.⁶⁹ The search for safe and effective haemoglobin-based oxygen carriers continues.⁷⁰^–⁷²

If blood loss has been massive or there are defects in the haemostatic system, specific component therapy may be indicated. Advances in patient retrieval, resuscitation protocols, techniques for rapid and real-time diagnosis, trauma teams and early ‘damage control’ surgery have improved the management of acutely haemorrhaging patients. Patients are now surviving with larger volumes of blood transfusion, but sepsis, acute lung injury and multiorgan failure remain major challenges, and blood transfusion is increasingly being recognised as a two-edged sword and probably a contributory factor to these complications in which microcirculatory dysfunction is recognised as central in the pathophysiology. As already discussed, the evidence that the immediate post-transfusion ability of stored red cells and haemoglobin to deliver and unload oxygen to microcirculation may be impaired is resulting in stored red cell concentrates not being automatically the first therapy for acutely enhancing oxygen delivery. Indeed, recent animal data indicate that the immediate clinical benefit of transfused red cells in treating hypovolaemic shock relates to reconstitution of the macrocirculation, with adverse effects on the functional capillary density in the microcirculation.⁷³

In previously healthy patients who have suffered an acute blood loss of less than 25% of their blood volume, restoring volume is more important than replacing oxygen-carrying capacity. Plasma volume expanders may preclude the necessity for allogeneic transfusion, especially if bleeding can be controlled. Clear fluids also allow time for transfusion compatibility testing. In the context of acute bleeding and hypovolaemic shock, the haemoglobin level is not the primary indicator for determining the need for allogeneic red cell transfusion. A normal human may survive a 30% deficit in blood volume without fluid replacement, in contrast to an 80% loss of red cell mass if normovolaemia is maintained. Minimisation of allogeneic blood transfusion is important and haemodilution with tolerance of anaemia is now accepted clinical practice. However, there are limits to haemodilution, not only from the oxygen-carrying capacity perspective. Marked reduction in blood viscosity is counterproductive as reactive peripheral vasoconstriction to maintain total peripheral resistance results in reduction in functional capillary density in the microcirculation. The role of clear fluids and the importance of plasma viscosity, colloid oncotic pressure, electrolyte composition and functional capillary density is the subject of current research.⁷⁴

A protocol approach to blood component therapy is generally not recommended, as each patient should be treated individually. However, this remains a controversial issue with advocates for ‘blind’, up-front component therapy with red cell and haemostatic components, especially fresh frozen plasma and cryoprecipitate or fibrinogen concentrates. With better understanding of coagulopathy in the critical haemorrhage setting and the importance of hypofibrinogenaemia and hyperfibrinolysis there is a reanalysis of the approach to blood component therapy.⁷⁵ In an elective situation in which normovolaemia is generally maintained from the outset, defects can be predicted and appropriate blood components prescribed, a protocol approach may be justified. This should be done in close consultation with the haematologist and hospital transfusion service.

Failure of haemostasis is common in critically bleeding patients and may be complex and multifactorial in pathogenesis.^76,⁷⁷ The importance of avoiding acidosis and hypothermia cannot be overemphasised as the triad of coagulopathy, acidosis and hypothermia has an extremely poor prognosis. Coagulopathy associated with critical haemorrhage is related to other aspects of the pathophysiology of the clinical setting. The assumption that there is a ‘generic coagulopathy’ vaguely referred to as DIC is incorrect and can result in inappropriate therapeutic decisions. The reader is referred to an expanding literature on this topic where it will be seen that there can be no standard approach to therapy.⁷⁷ This author is increasingly convinced that the pathophysiology of coagulopathy, when occurring in the context of critical haemorrhage, should be viewed as related to the primary insult/initiating event, and that a ‘secondary’ coagulopathy (e.g. massive stored blood transfusion, haemodilution, hypothermia, continuing tissue hypoxia, etc.) may compound the problem in the resuscitated patient. The primary mechanisms of coagulopathy relating to the initiating event may relate to trauma, hypoxia, pregnancy, microbial infection, envenomation or iatrogenic (antithrombotic) agents. In all circumstances there is activation or inhibition of some aspect of the haemostatic system, and therapy is better informed if the varied mechanisms are understood. Frequently, complex tests are required for definitive diagnosis, but the urgency of the situation cannot always await the results, and therapy may be initiated on clinical evidence with minimal laboratory information.

Many trauma patients have coagulopathy at presentation that is more related to hypovolaemic shock and not consumption or dilution. Recent evidence indicates that excessive activation of the protein C system and hypofibrinogenaemia secondary to secondary hyperfibrinolysis are important.⁷⁸ Except for severe abnormalities, haemostatic laboratory parameters correlate poorly with clinical evidence of haemostatic failure. In the massively transfused patient thrombocytopenia and impaired platelet function are the most consistent significant haematological abnormalities, correction of which may be associated with control of microvascular bleeding. Coagulation deficiency from massive blood loss is initially confined to factors V and VIII. Activated partial thromboplastin time (APTT), prothrombin time (PT) and fibrinogen assay should be performed, but the urgency of the situation does not usually allow for other specific factor assays.

A problem with the standard screen tests of coagulation function is that they do not give information about the actual formation of the haemostatic plug, its size, structure or stability. For this reason there is increasing interest in global tests of haemostatic plug formation and stability such as thromboelastography, thrombin generation tests and clot waveform analysis in which changes in light transmission in routine APTT are measured.⁷⁹ Fresh frozen plasma should be infused if the test results are abnormal, including cryoprecipitate or fibrinogen concentrates if the fibrinogen level is < 1 g/l. A case can be made for prophylactic fresh frozen plasma in infusions in patients with massive blood loss which has been replaced with red cell concentrates and plasma substitutes.

With ongoing bleeding with associated microvascular oozing various approaches may be taken. Having ensured that all identifiable haemostatic defects have been corrected, questions arise as to the role of fresh blood and, more recently, recombinant activated factor VII (rFVIIa).⁸⁰ The use of freshly collected whole blood remains controversial and is an ‘emotive’ subject needing some comment. The provision of fresh whole blood can present logistic, ethical and safety problems needing consideration. Many transfusion medicine specialists categorically state that there is never an indication for fresh whole blood. Such dogma can be difficult to defend in the clinical setting of the massive haemorrhage and transfusion syndrome where pathophysiology remains poorly understood and specific blood component therapy may be ineffective.

The following statements reasonably summarise the current status of fresh blood transfusion:

• Fresh blood provides immediately functioning oxygen-carrying capacity, volume and haemostatic factors in the one product

• The number of allogeneic donors to whom a patient is exposed can be reduced

• Problems associated with infusion of massive volumes of stored blood can be minimised/avoided

• The risk of transfusion-related viral infections is possibly higher than fully tested and stored blood

• The benefit in controlling haemorrhage probably relates to the presence of immediately functioning platelets