Introduction

Anaemia is a condition in which the absolute number of red cells in the circulation is abnormally low. The diagnosis is usually made on the basis of the full blood count (FBC). This, together with the blood film, offers qualitative as well as quantitative data on the blood components, and a set of normal values is shown in Table 13.1.1.

Table 13.1.1 Full blood count: normal parameters

Haemoglobin (Hb)
Males	13.5–18 g/dL
Females	11.5–16.5 g/dL
Red blood cell count
Males	4500–6500 × 109/L
Females	3900–5600 × 109/L
Haematocrit
Males	42–54%
Females	37–47%
MCH	27–32 pg
MCHC	32–36 g/dL
MCV	76–98 fL
Reticulocytes	0.2–2%
White blood cells	4–11 × 109/L
Neutrophils	1.8–8 × 109/L
Eosinophils	0–0.6 × 109/L
Basophils	0–0.2 × 109/L
Lymphocytes	1–5 × 109/L
Monocytes	0–0.8 × 109/L
Platelets	150–400 × 109/L

MCH, Hb divided by RBC; MCHC, Hb divided by HCT; MCV, HCT divided by RBC.

Most automated counting machines now give the red cell distribution width (RDW), a measure of degree of variation of cell size.

The average lifespan of a normal red blood cell in the circulation is from 100 to 120 days. Aged red cells are removed by the reticuloendothelial system, but under normal conditions are replaced by the marrow such that a dynamic equilibrium is maintained. Anaemia develops when red cell loss exceeds red cell production. It follows that the anaemic patient is doing at least one of three things: not producing enough red cells, destroying them too quickly or bleeding.

The overriding functional importance of the red cell resides in its ability to transport oxygen, bound to the haemoglobin molecule, from the lungs to the tissues. Functionally, anaemia may be regarded as an impairment in the supply of oxygen to the tissues and the adverse effects of anaemia, from whatever cause, are a consequence of the resultant tissue hypoxia. Anaemia is not a diagnosis: rather, it is a clinical or a laboratory finding that should prompt the search for an underlying cause (Table 13.1.2).

Table 13.1.2 Causes of anaemia

Haemorrhage

Traumatic

Non-traumatic

Acute or chronic

Production defect

Megaloblastic anaemia

Vitamin B12 deficiency

Folate deficiency

Aplastic anaemia

Pure red cell aplasia

Myelodysplastic syndromes

Invasive marrow diseases

Chronic renal failure

Decreased RBC survival (haemolytic anaemia)

Congenital

Spherocytosis

Elliptocytosis

Glucose-6-phosphate-dehydrogenase deficiency

Pyruvate kinase deficiency

Haemoglobinopathies: sickle cell diseases

Acquired autoimmune haemolytic anaemia, warm

Acquired autoimmune haemolytic anaemia, cold

Microangiopathic haemolytic anaemias

RBC mechanical trauma

Infections

Paroxysmal nocturnal haemoglobinuria

RBC, red blood cell.

ANAEMIA SECONDARY TO HAEMORRHAGE

Clinical features

While it may be obvious on history and examination that a patient is bleeding, occasionally the source of blood loss is occult and the extent of loss underestimated.

In the context of trauma the history often gives clear pointers to both sites and extent of blood loss. Consideration of the mechanism of injury may allow anticipation of occult pelvic, intraperitoneal or retroperitoneal bleeding. Intracranial bleeding is never an explanation for hypovolaemic shock in an adult. In the context of non-trauma it is essential to obtain an obstetric and gynaecological history in women of childbearing age. The remainder of the formal history may supply information essential in determining the aetiology of anaemia. The past medical history may point to a known haematological abnormality or to a chronic disease process. A drug and allergy history is always relevant. Many drugs cause marrow suppression, haemolytic anaemia and bleeding. The family history points to hereditary disease; the social history may alert the clinician to an unusual occupational exposure in the patient’s past or, more likely, to recreational activities liable to exacerbate an ongoing disease process. The systems review is particularly relevant to the consultation with middle-aged or elderly male patients, who must be asked about symptoms of altered bowel habit and weight loss.

The symptomatology of anaemia proceeds from vague complaints of tiredness, lethargy and impaired performance through to more sharply defined entities such as shortness of breath on exertion, giddiness, restlessness, apprehension, confusion, and collapse. Comorbid conditions may be exacerbated (the dyspnoea of chronic obstructive airway disease) and occult pathologies unmasked (exertional angina in ischaemic heart disease).

Anaemia of insidious onset is generally better tolerated than that of rapid onset because of cardiovascular and other compensatory mechanisms. Acute loss of 40% of the blood volume may result in collapse, whereas in certain developing countries it is not rare for patients with haemoglobin concentrations 10% of normal to be ambulant. Trauma superimposed on an already established anaemia can lead to rapid decompensation.

The cardinal sign of anaemia is pallor. This can be seen in the skin, the lips, the mucous membranes and the conjunctival reflections. Yet not all anaemic patients are pallid, and not all patients with a pale complexion are anaemic. Patients who have suffered an acute haemorrhage may show evidence of hypovolaemia: tachycardia, hypotension, cold peripheries and sluggish capillary refill. The detection of postural hypotension is an important pointer towards occult blood loss. Conversely, patients with anaemia of insidious onset are not hypovolaemic and may manifest high-output cardiac failure as a physiological response to hypoxia.

Other features of the physical examination may provide clues to the aetiology of anaemia. The glossitis, angular stomatitis, koilonychia and oesophageal web of iron-deficiency anaemia are uncommon findings. Bone tenderness, lymphadenopathy, hepatomegaly and splenomegaly may point to an underlying haematological abnormality. The rectal and gynaecological examinations can sometimes be diagnostic.

Clinical investigations

The full blood count often reveals an anaemia that has not been clinically suspected and that must be interpreted in the light of the history and examination. If the anaemia is mild it may be a chance finding with little relevance to the patient’s presenting complaint, but such a finding should never be ignored. At the very least a follow-up blood count should be arranged.

Anaemic patients have a low red cell count, a low haematocrit and a low haemoglobin, but some caveats need to be borne in mind:

Red cell morphology, particularly the mean corpuscular volume (MCV), can help elucidate the cause of anaemia. The finding of a pancytopenia suggests a problem in haematopoiesis, rather than haemolysis or blood loss. In women of childbearing age, assay of blood or urine β-HCG is important.

Treatment

The principles of management of haemorrhage are as follows:

The indications for red cell transfusion are discussed in Chapter 13.5. The faster the onset of the anaemia, the greater the need for urgent replacement. Patients who are tolerating their anaemia may require no more than an appropriate diet with or without the addition of haematinics. Elderly patients with severe bleeding often need red cells urgently. Excessive administration of colloid and/or crystalloid precipitates left ventricular failure, and it can then be difficult to administer red cells.

Chronic haemorrhage

The finding of a hypochromic microcytic anaemia on blood film is usually indicative of iron deficiency and, in the absence of an overt history of bleeding, should prompt the search for occult blood loss. Iron-deficiency anaemia may be due to malnutrition, but inadequate dietary intake of iron is not usually the sole cause of anaemia in developed countries: much more commonly it is the result of chronic blood loss from the gastrointestinal (GI) tract, the uterus or the renal tract. More unusual causes are haemoptysis and recurrent epistaxes.

Patients present with insidious and rather vague symptoms. They may be unaware that they are bleeding and will probably show none of the trophic skin, nail and mucosal changes of iron deficiency. The automated cell count, in addition to showing a hypochromic, microcytic picture, may also show a raised red cell distribution width, which reflects anisocytosis on the blood film.

Iron studies may confirm the diagnosis of iron deficiency without pointing to the underlying cause. Serum iron and ferritin are low and total iron-binding capacity is high.

Disposition

If the source of blood loss is obvious, for example heavy menstrual bleeding, then appropriate referral may be all that is indicated. If the source is not obvious, particularly in older patients, then sequential investigation of the GI tract and the renal tract may be indicated. Decisions to admit or discharge these patients depend on the red cell reserves, the patient’s cardiorespiratory status, home circumstances and the likelihood of compliance with follow-up.

The anaemia itself can be corrected with oral iron supplements: 200 mg of ferrous sulphate three times daily is an appropriate regimen, although single daily doses are often more acceptable to the patient and have fewer GI side effects.

ANAEMIA SECONDARY TO DECREASED RED CELL PRODUCTION

Other causes of decreased red cell production

Bone marrow failure is rarely encountered in emergency medicine practice. The physician must be alert to the unusual, insidious or sinister presentation, and be particularly attuned to the triad of decreased tissue oxygenation, immunocompromise and a bleeding diathesis that may herald a pancytopenia. An FBC may dictate the need for haematological consultation, hospital admission and further investigation.

Among the entities to be considered are the aplastic anaemias, characterized by a pancytopenia secondary to failure of pluripotent myeloid stem cells. Half of cases are idiopathic, but important aetiologies are infections (e.g. non-A, non-B hepatitis), inherited diseases (e.g. Fanconi’s anaemia), irradiation, therapeutic or otherwise, and – most important in the emergency setting – drugs. Drugs that have been implicated in the development of aplastic anaemia include, in addition to antimetabolites and alkylating agents, chloramphenicol, chlorpromazine and streptomycin.

Characteristic of patients with a primary marrow failure is the absence of splenomegaly and the absence of a reticulocyte response. There is a correlation between prognosis and the severity of the pancytopenia. Platelet counts less than 20 × 10⁹/L and neutrophil counts less than 500/mL equate to severe disease. Depending on the severity of the accompanying anaemia, patients may require red cell and sometimes platelet transfusion in the ED, as well as broad-spectrum antibiotic cover. It is imperative to stop all medications that might be causing the marrow failure. Other forms of marrow failure include pure red cell aplasia, where marrow red cell precursors are absent or diminished. This can be a complication of haemolytic states in which a viral insult leads to an aplastic crisis (see haemolytic anaemias).

The myelodysplastic syndromes are a group of disorders primarily affecting the elderly. In these states there is no reduction in marrow cellularity but the mature red cells, granulocytes and platelets generated from an abnormal clone of stem cells are disordered and dysfunctional. There is peripheral pancytopenia. These disorders are classified according to observed cellular morphology (Table 13.1.5). These conditions were once termed ‘preleukaemia’, and one-third of patients progress to acute myeloid leukaemia.

Table 13.1.5 Classification of the myelodysplastic syndromes

Refractory anaemia

Refractory anaemia with ringed sideroblasts

Refractory anaemia with excess of blasts

Chronic myelomonocytic leukaemia

Two more causes of failure of erythropoiesis might be mentioned. One is due to invasion of the marrow and disruption of its architecture by extraneous tissue, the commonest cause being metastatic cancer. Finally, but not at all uncommon, is the anaemia of chronic renal failure, where deficient erythropoiesis is attributed to decreased production of erythropoietin. Most patients with chronic renal failure on dialysis treatment tolerate a moderate degree of anaemia, but occasionally require either transfusion or treatment with erythropoietin. Emergency physicians should recognize anaemia as a predictable entity in patients with chronic renal failure, usually not requiring any action.

ANAEMIA SECONDARY TO DECREASED RED CELL SURVIVAL: THE HAEMOLYTIC ANAEMIAS

Patients whose main problem is haemolysis are encountered rarely in the ED. The most fulminant haemolytic emergency one could envisage is that following transfusion of ABO-incompatible blood (discussed in Ch. 13.5), a vanishingly rare event where proper procedures are followed. Haemolysis and haemolytic anaemia are occasionally encountered in decompensating patients with multisystem problems. Rarely, first presentations of unusual haematological conditions occur.

Some of the haemolytic anaemias are hereditary conditions in which the inherited disorder is an abnormality intrinsic to the red cell, its membrane, its metabolic pathways or the structure of the haemoglobin contained in the cells. Such red cells are liable to be dysfunctional, and to have increased fragility and a shortened lifespan. Lysis in the circulation may lead to clinical jaundice as bilirubin is formed from the breakdown of haemoglobin. Lysis in the reticuloendothelial system generally does not cause jaundice but may produce splenomegaly. The anaemia tends to be normochromic normocytic; sometimes a mildly raised MCV is due to an appropriate reticulocyte response from a normally functioning marrow. Serum bilirubin may be raised even in the absence of jaundice. Urinary urobilinogen and faecal stercobilinogen are detectable and serum haptoglobin is depleted. The antiglobulin (Coombs’) test is important in the elucidation of some haemolytic anaemias. In this test, red cells coated in vivo (direct test) or in vitro (indirect test) with IgG antibodies are washed to remove unbound antibodies, then incubated with an antihuman globulin reagent. The resultant agglutination is a positive test.

Any chronic haemolytic process may be complicated by an ‘aplastic crisis’. This is a usually transient marrow suppression brought on by a viral infection which can result in a severe and life-threatening anaemia. Red cell transfusion in these circumstances may be life-saving.

Hereditary spherocytosis

A deficiency of the red cell wall protein, spectrin, leads to loss of deformability and increased red cell fragility. These cells are destroyed prematurely in the spleen. The condition may present at any age, with anaemia, intermittent jaundice and cholelithiasis. Patients are Coombs’ negative and show normal red cell osmotic fragility. Splenectomy radically improves general health. Hereditary elliptocytosis is a similar disease, with usually a milder course.

Glucose-6-phosphate dehydrogenase deficiency

Glucose-6-phosphate dehydrogenase (G6PD) generates reduced glutathione, which protects the red cell from oxidant stress. G6PD deficiency is an X-linked disorder present in heterozygous males and homozygous females. The disorder is commonly seen in West Africa, southern Europe, the Middle East and South East Asia. Oxidant stress leads to severe haemolytic anaemia. Precipitants include fava beans, antimalarial and analgesic drugs, and infections. The enzyme deficiency can be demonstrated by direct assay, and treatment is supportive.

Sickle cell anaemia

Whereas in the thalassaemias there is a deficiency in a given globin chain within the haemoglobin (Hb) molecule, in the haemoglobinopathies a given globin chain is present but structurally abnormal. HbS differs from normal HbA by one amino acid residue: valine replaces glutamic acid at the sixth amino acid from the N-terminus of the β-globin chain. Red cells containing HbS tend to ‘sickle’ at states of low oxygen tension. The deformed sickle-shaped red cell has increased rigidity, which causes it to lodge in the microcirculation and sequester in the reticuloendothelial system – the cause of a haemolytic anaemia.

Sickle cell disease is encountered in Afro-Caribbean people. The higher incidence in tropical areas is attributed to the survival value of the β-S gene against falciparum malaria. Heterozygous individuals have ‘sickle trait’ and are usually asymptomatic. Homozygous (HbSS) individuals manifest the disease in varying degrees. The haemolytic anaemia is usually in the range of 60–100 g/L and can be well tolerated because HbS offloads oxygen to the tissues more efficiently than HbA.

A patient with sickle cell disease may occasionally develop a rapidly worsening anaemia. This may be due to:

In any of these circumstances transfusion may be life-saving. However, these events are unusual and more commonly encountered is the vaso-occlusive crisis. A stressor – for example infection, dehydration, or cold – causes sickle cells to lodge in the microcirculation. Bone marrow infarction is one well-recognized complication of the phenomenon, but virtually any body system can be affected. Common presenting complaints include acute spinal pain, abdominal pain (the mesenteric occlusion of ‘girdle sequestration’), chest pain (pulmonary vascular occlusion), joint pain, fever (secondary to tissue necrosis), neurological involvement (translent ischaemic attacks, strokes, seizures, obtundation, coma), respiratory embarrassment and hypoxia, priapism, ‘hand-foot syndrome’ (dactylitis of infancy), haematuria (nephrotic syndrome, papillary necrosis), skin ulcers of the lower limbs, retinopathies, glaucoma and gallstones.

Most patients presenting with a vaso-occlusive crisis know they have the disease but otherwise the differential diagnosis is difficult. Sickle cells may be seen on the blood film, and can also be induced by deoxygenating the sample. Hb electrophoresis can establish the type of Hb present. Other investigations are dictated by the presentation, and may include blood cultures, urinalysis and culture, chest X-ray, arterial blood gases and electrocardiograph.

Pain relief should commence early. A morphine infusion may be required for patients with severe ongoing pain. Other supportive measures are dictated by the presentation. Intravenous fluids are particularly important for patients with renal involvement. Aim to establish a urine output in excess of 100 mL/h in adults. Antibiotic cover may be required in the case of febrile patients with lung involvement. It may be impossible to differentiate between pulmonary vaso-occlusion and pneumonia. Many patients with sickle cell disease are effectively splenectomized owing to chronic splenic sequestration with infarction, and are prone to infection from encapsulated bacteria. The choice of antibiotic depends on the clinical presentation. Indications for exchange transfusion are shown in Table 13.1.6. The efficacy of exchange transfusion in painful crises remains unproven.

Table 13.1.6 Indications for exchange transfusion in sickle cell crisis

Neurological presentations: TIAs, stroke, seizures

Lung involvement (PaO2 < 65 mmHg with FiO2 60%)

Sequestration syndromes

Priapism

TIA, transient ischaemic attack

Thalassaemias

There is a high incidence of β-thalassaemia trait among people of Mediterranean origin, although in fact the region of high frequency extends in a broad band east to South East Asia.

Thalassaemias are disorders of haemoglobin synthesis. In the haemoglobin molecule, four haem molecules are attached to four long polypeptide globin chains. Four globin chain types (each with their own minor variations in amino acid order) are designated α, β, γ and δ. Haemoglobin A comprises two α and two β chains; 97% of adult haemoglobin is HbA. In thalassaemia there is diminished or absent production of either the α chain (α-thalassaemia) or the β chain (β-thalassaemia). Most patients are heterozygous and have a mild asymptomatic anaemia, although the red cells are small. In fact, the finding of a marked microcytosis in conjunction with a mild anaemia suggests the diagnosis.

There are four genes on paired chromosome 16 coding for α-globin and two genes on paired chromosomes 11 coding for β-globin. α-Thalassaemias are associated with patterns of gene deletion as follows: (-/-) is Hb-Barts hydrops syndrome, incompatible with life, and (-α/-) is HbH disease.

Patients who are heterozygous for β-thalassaemia have β-thalassaemia minor or thalassaemia trait. They are usually symptomless. Homozygous patients have β major.

Diagnosis of the major clinical syndromes is usually possible through consideration of the presenting features in conjunction with an FBC, blood film and Hb electrophoresis.

HbH disease patients present with moderate haemolytic anaemia and splenomegaly. The HbH molecule is detectable on electrophoresis and comprises unstable β tetramers. α Trait occurs with deletion of one or two genes. Hb, MCV and mean corpuscular haemoglobin(MCH) are low, but the patient is often asymptomatic.

β major becomes apparent in the first 6 months of life with the decline of fetal Hb. There is a severe haemolytic anaemia, ineffective erythropoiesis, hepatosplenomegaly and failure to thrive. With improved care many of these patients survive to adulthood, and may possibly present to the ED, where transfusion could be life-saving. Patients with β trait may be encountered in the ED relatively frequently. They are generally asymptomatic, with a mild hypochromic microcytic anaemia. It is important not to work these patients up continually for iron deficiency, and not to subject them to inappropriate haematinic therapy.

Acquired haemolytic anaemias

Many of the acquired haemolytic anaemias are autoimmune in nature, a manifestation of a type II (cytotoxic) hypersensitivity reaction. Here, normal red cells are attacked by aberrant autoantibodies targeting antigens on the red cell membrane. These reactions may occur more readily at 37 °C (warm autoimmune haemolytic anaemia, or AIHA), or at 4 °C (cold AIHA). Warm AIHA is more common. Red cells are coated with IgG, complement or both. The cells are destroyed in the reticuloendothelial system. Fifty per cent of cases are idiopathic, but other recognized causes include lymphoproliferative disorders, neoplasms, connective tissue disorders, infections and drugs (notably methyldopa and penicillin). Patients have haemolytic anaemia, splenomegaly and a positive Coombs’ test. In the ED setting it is important to stop any potentially offending drugs and search for the underlying disease. The idiopathic group may respond to steroids, other immunosuppressive or cytotoxic drugs or splenectomy.

In cold AIHA, IgM attaches to the I red cell antigen in the cooler peripheries. Primary cold antibody AIHA is known as cold haemagglutinin disease. Other causes include lymphoproliferative disorders, infections such as mycoplasma, and paroxysmal cold haemoglobinuria. Patients sometimes manifest Reynaud’s disease and other manifestations of circulatory obstruction. Symptoms worsen in winter. Red cell lysis leads to haemoglobinuria.

Microangiopathic haemolytic anaemia

In this important group of conditions intravascular haemolysis occurs in conjunction with a disorder of microcirculation. Important causes are shown in Table 13.1.7.

Table 13.1.7 Causes of microangiopathic haemolytic anaemia

Disseminated intravascular coagulation

Haemolytic uraemic syndrome

HELLP

Malignancy

Malignant hypertension

Snake envenoming

Thrombotic thrombocytopenic purpura

Vasculitis

Haemolytic uraemic syndrome and thrombotic thrombocytopenic purpura

These are probably manifestations of the same pathological entity, with haemolytic uraemic syndrome occurring in children and thrombotic thrombocytopenic purpura most commonly in the fourth decade, especially in women. The primary lesion is likely to be in the vascular endothelium. Fibrin and platelet microthrombi are laid down in arterioles and capillaries, possibly as an autoimmune reaction. The clotting system is not activated. Haemolytic anaemia, thrombocytopenia and acute renal failure are sometimes accompanied by fever and neurological deficits.

In adults, the presentation is usually one of a neurological disturbance (headache, confusion, obtundation, seizures or focal signs). The blood film reveals anaemia, thrombocytopenia, reticulocytosis and schistocytes. Coombs’ test is negative.

Patients require hospital admission. Adults with this condition may require aggressive therapy with prednisone, antiplatelet therapy, further immunosuppressive therapy and plasma exchange transfusions.

HELLP syndrome

HELLP stands for haemolysis, elevated liver enzymes and a low platelet count, and is seen in pregnant women in the context of pre-eclampsia. Treatment is as for pre-eclampsia, early delivery of the baby being of paramount importance.

Disseminated intravascular coagulation

The introduction of procoagulants into the circulation resulting in the overwhelming of anticoagulant control systems may occur as a consequence of a substantial number of pathophysiological insults – obstetric, infective, malignant and traumatic. Disseminated intravascular coagulation has an intimate association with shock, from any cause. The widespread production of thrombin leads to deposition of microthrombi, bleeding secondary to thrombocytopenia and a consumption coagulopathy, and red cell damage within abnormal vasculature leading to a haemolytic anaemia.

Recognition of this condition prompts intensive care admission and aggressive therapy. Principles of treatment include definitive management of the underlying cause and, from the haematological point of view, replacement therapy that may involve transfusion of red cells, platelets, FFP, and cryoprecipitate. There may be a role for heparin and other anticoagulant treatments if specific tissue and organ survival is threatened by thrombus.

Paroxysmal noctural haemoglobinuria

This entity is unusual in that an intrinsic red cell defect is seen in the context of an acquired haemolytic anaemia. A somatic stem cell mutation results in a clonal disorder. A family of membrane proteins (CD55, CD59 and C8 binding protein) is deficient and renders cells prone to complement-mediated lysis. Because the same proteins are deficient in white cells and platelets, in addition to being anaemic patients are prone to infections and haemostatic abnormalities. They may go on to develop aplastic anaemia or leukaemia. Treatment is supportive. Marrow transplant can be curative.

Other causes of haemolysis

Haemolysis may be due to mechanical trauma, as in ‘March haemoglobinuria’. Artificial heart valves can potentially traumatize red cells. Historically, ball-and-cage type valves have been most prone to cause haemolysis, whereas disc valves are more thrombogenic. Improvements in design have made cardiac haemolytic anaemia very rare. Haemolysis is sometimes seen in association with a number of infectious diseases, notably malaria. Other infections that have been implicated are listed in Table 13.1.8. Certain drugs and toxins are associated with haemolytic anaemia (Table 13.1.9). The haemolytic anaemia that is commonly seen in patients with severe burns is attributed to direct damage to the red cells by heat.

Table 13.1.8 Infections associated with haemolysis

Babesiosis

Bartonella

Clostridia

Cytomegalovirus

Coxsackie virus

Epstein-Barr virus

Haemophilus

Herpes simplex

HIV

Malaria, especially Plasmodium falciparum (Blackwater fever)

Measles

Mycoplasma

Varicella

Table 13.1.9 Drugs and toxins associated with haemolysis

Antimalarials

Arsine (arsenic hydride)

Bites: bees, wasps, spiders, snakes

Copper toxicity

Dapsone

Lead (plumbism)

Local anaesthetics: lidocaine, benzocaine

Nitrates, nitrites

Sulfonamides

Introduction

Neutropenia is defined as a decrease in the number of circulating neutrophils. The neutrophil count varies with age, sex and racial grouping. The severity of neutropenia is usually graded as follows:¹

The risk of infection rises as the neutrophil count falls and becomes significant once the neutrophil count drops below 1.0 × 10⁹/L. Febrile neutropenia is a term used to describe the clinical scenario where a patient has a neutrophil count less than 1.0 × 10⁹/L in the presence of a temperature greater than or equal to 38 °C or if the patient is systemically unwell with signs or symptoms of sepsis. Neutropenic patients are at greater risk of overwhelming infection if the onset of the neutropenia is acute rather than chronic and, in the case of patients receiving cancer chemotherapy, if the absolute neutrophil count is in the process of falling rather than rising.

Signs or symptoms of infection in the presence of severe neutropenia constitute a true emergency that mandates rapid assessment and aggressive management to prevent progression to overwhelming sepsis. In the emergency department (ED) setting this is most commonly encountered when a patient presents with fever in the context of chemotherapy for cancer.

Pathophysiology and aetiology

Polymorphonuclear neutrophils are formed in marrow from the myelogenous cell series. Pluripotent haematopoietic stem cells are committed to a particular cell lineage through the formation of colony-forming units, which further differentiate to form given white cell precursors. The mature neutrophil has a multilobed nucleus and granules in the cytoplasm. The cells are termed ‘neutrophilic’ because of the lilac colour of the granules caused by the uptake of both acidic and basic dyes.

The neutrophils leave the marrow and enter the circulation, where they have a lifespan of only 6–10 h before entering the tissues. Here they migrate by chemotaxis to sites of infection and injury, and then phagocytose and destroy foreign material. In health, about half of the available mature neutrophils are in the circulation. ‘Marginal’ cells are adherent to vascular endothelium or in the tissues and are not measured by the full blood count. Some individuals have fixed increased marginal neutrophil pools and decreased circulating pools; they are said to have benign idiopathic neutropenia.

For a previously normal individual to become neutropenic there must be decreased production of neutrophils in the marrow, decreased survival of mature neutrophils or a redistribution of neutrophils from the circulating pool. The important causes are shown in Table 13.2.1.

Table 13.2.1 Important causes of neutropenia

Decreased production

Aplastic anaemia

Leukaemias

Lymphomas

Metastatic cancer

Drug-induced agranulocytosis

Megaloblastic anaemias Vitamin B12 deficiency Folate deficiency

CD8 and large granular lymphocytosis Myelodysplastic syndromes Decreased survival Idiopathic immune related Systemic lupus erythematosis Felty syndrome Drugs Redistribution Sequestration (hypersplenism) Increased utilization (overwhelming sepsis) Viraemia

It is a defect in neutrophil production that is most likely to prove life threatening. Consumption of neutrophils in the periphery, as occurs early in infectious processes, is likely to be rapidly compensated for by a functioning marrow. Fortunately, most of the primary diseases of haematopoiesis are rare, and in practice many of the acquired neutropenias are drug induced. Processes interfering with haematopoiesis, often involving autoimmune mechanisms, may affect neutrophils both in the marrow and in the periphery. Some drugs cause neutropenia universally but many more reactions are idiosyncratic, be they dose-related or independent of dose. Some commonly implicated drugs are listed in Table 13.2.2. Cancer chemotherapy drugs are now recognized as the commonest cause of neutropenia.

Table 13.2.2 Drugs commonly associated with neutropenia

Antibiotics: chloramphenicol, sulfonamides, isoniazid, rifampicin, β-lactams, carbenicillin

Antidysrhythmic agents: quinidine, procainamide

Antiepileptics: phenytoin, carbamazepine

Antihypertensives: thiazides, ethacrynic acid, captopril, methyldopa, hydralazine

Antithyroid agents

Chemotherapeutic agents: especially methotrexate, cytosine arabinoside, 5-azacytidine, azothioprine, doxorubicin, daunorubicin, hydroxyurea, alkylating agents

Connective tissue disorder agents: phenylbutazone, penicillamine, gold

H2-receptor antagonists

Phenothiazines, especially chlorpromazine

Miscellaneous: imipramine, allopurinol, clozapine, ticlopidine, tolbutamide

Clinical features

Neutropenia is frequently anticipated based on the clinical presentation, such as fever developing in the context of cancer chemotherapy, by far the most common scenario in which severe neutropenia is seen in the ED. Alternatively, it may be identified in the course of investigation for a likely infective illness, or it might be an incidental finding during investigation for an unrelated condition.

Chronic neutropenia may be asymptomatic unless secondary or recurrent infections develop. Acute severe neutropenia may present with fever, sore throat, and mucosal ulceration or inflammation.² Symptoms or signs of an associated disease process may also be present, such as pallor from anaemia, or bleeding from thrombocytopenia, as might occur in conditions causing pancytopenia.

The history of the mode of onset and duration of the illness is important. Systems enquiry may reveal cough, headache and photophobia, a diarrhoeal illness, or urinary symptoms. The past history may reveal a known haematological illness or previous evidence of immunosuppression, such as frequent and recurrent infections. A detailed drug history is vital. Most neutropenic drug reactions occur within the first 3 months of taking a drug.

In the ED, vital signs, including pulse, blood pressure, temperature, respiratory rate and pulse oximetry, should be performed at initial assessment and monitored regularly until disposition. Attention should be paid to identifying early signs of severe sepsis and the progression to septic shock.

Physical examination may reveal necrotizing mucosal lesions, pallor, petechial rashes, lymphadenopathy, bone tenderness, abnormal tonsillar or respiratory findings, spleno- or other organomegaly.² Careful examination of the skin of the back, the lower limbs and the perineum for evidence of infection is important. The presence of indwelling venous access devices should be noted and insertion sites inspected for evidence of inflammation or infection.

Clinical investigation

Investigation in the ED is firstly aimed at confirming and quantifying the severity of neutropenia, identifying the cause, and then at identifying the focus and severity of infection. An urgent full blood count and blood film should be ordered in any patient who is suspected of suffering febrile neutropenia. A coagulation profile and biochemistry, including electrolytes and creatinine, glucose and liver function tests may be indicated once severe neutropenia is confirmed. Anaemic patients may require a group-and-hold or cross-match.

Microbiological cultures aimed at isolating a causative organism should be taken but antibiotics should not be unreasonably delayed in the presence of fever and confirmed significant neutropenia. Blood cultures should be taken at the time of cannulation and if possible prior to the instigation of antibiotic therapy. Throat swab, swabs of skin lesions and indwelling venous access device sites, urinalysis and urine culture may be indicated depending on the clinical picture. Patients with apparent central nervous system infections might require a lumbar puncture, but this should be postponed or even cancelled in the presence of an uncorrected coagulopathy, signs of raised intracranial pressure, focal neurological signs, or haemodynamic instability. Antibiotics, if clinically indicated, should be commenced prior to lumbar puncture.

Treatment

Management of the patient with confirmed febrile neutropenia in the ED involves early recognition and treatment of bacterial infection, and institution of supportive care to prevent progression to overwhelming sepsis and shock. Evolving or established haemodynamic instability requires immediate, aggressive resuscitation.

Empiric broad-spectrum antibiotic therapy should be started in the ED after drawing blood for culture in any patient with fever and confirmed significant neutropenia. This strategy has played a pivotal role in reducing mortality rates in febrile neutropenia.³ There is no clear consensus approach to which particular empiric antibiotic regime should be used. Practices vary widely amongst institutions and regions, and are influenced by local patterns of infection, prevalence and risk of inducing resistant organisms, and acquisition costs.⁴ In general antibiotics should provide good cover for both Gram-positive and Gram-negative organisms. With increased use of indwelling venous access devices for cancer chemotherapy, there has been an increase in the incidence of sepsis due to Gram-positive organisms such as coagulase negative staphylococci, S. aureus and MRSA.⁵ Although occurring infrequently, bacteraemia due to Pseudomonas aeruginosa is associated with a high morbidity and mortality and therefore should also be covered.⁶ A reasonable initial regime might include ticarcillin/clavulanate plus either a cephalosporin, such as ceftazidime, or an aminoglycoside, such as gentamicin. The addition of vancomycin might be considered if the patient is in shock, is known to be colonized with MRSA or has clinical evidence of a catheter-related infection in a unit with a high incidence of MRSA. Empiric antifungal therapy is not generally required unless there is persistent fever in high-risk patients beyond 96 h of antibacterial therapy.⁶

Disposition

The presence of significant neutropenia with fever generally mandates admission to hospital. Patients with severe acute neutropenia without an established aetiology will also generally require admission regardless of the presence or absence of fever. Both the haematological abnormality and the likely presence of infection require investigation. Sometimes the aetiology of the neutropenia will be evident; on other occasions marrow aspiration and biopsy will be required.

There is emerging evidence that a subset of febrile neutropenic patients can be identified who are at low risk of life-threatening complications and in whom duration of hospitalization and intensity of treatment may be safely reduced.⁷ Strategies that involve outpatient treatment of low-risk patients with oral antibiotics have also been evaluated.³ Such regimens are reliant upon accurate prediction of risk, as well as the availability of structured programmes and resources, and are not yet in widespread use.

Prognosis

The prognosis of the neutropenic patient is largely dependent upon the underlying aetiology of the condition. Febrile neutropenia has in the past been associated with a significant mortality rate which varies depending on the organism causing the infection. Improvements in therapy, such as rapid treatment with empiric broad-spectrum antibiotics, have significantly reduced mortality rates from this condition. Overall mortality rates for patients with febrile neutropenia have reduced from more than 20% to less than 4% in recent data sets.^3,7

Introduction

Thrombocytopenia is defined as a reduction in the number of circulating platelets, the normal circulating platelet count being 150–400 × 10⁹/L. It is the most common cause of abnormal bleeding.¹ Like anaemia, thrombocytopenia itself is not a diagnosis, but rather a manifestation of another underlying disease process.

In the emergency department setting, thrombocytopenia may present as an incidental finding on a routine blood count or may be diagnosed in the context of abnormal bleeding. In most cases, the underlying aetiology can be determined by a careful history and physical examination combined with interpretation of the blood count.

Aetiology

The clinically important causes of thrombocytopenia are outlined in Table 13.3.1. Diagnoses are classified by pathological process. It should be noted that more than one pathological process may be present.

Table 13.3.1 Causes of thrombocytopenia

Pseudothrombocytopenia
Platelet clumping Collection into anticoagulant (EDTA) Platelet agglutinins Giant platelets

Increased platelet destruction

Immune Primary Idiopathic thrombocytopenic purpura (ITP) Secondary Autoimmune thrombocytopenia associated with other disorders Graves’ disease, Hashimoto’s thyroiditis, systemic lupus erythematosus HIV-related thrombocytopenia Drug-induced thrombocytopenia Heparin, gold salts, quinine/quinidine, sulfonamides, rifampicin, H2-blockers, indometacin, carbamazepine, valproic acid, ticlopidine, clopidogrel, monoclonal antibodies (infliximab, efalizumab, rituximab) Post-transfusion purpura Non-immune Thrombotic thrombocytopenic purpura – haemolytic uraemic syndrome Pregnancy Gestational benign thrombocytopenia Pre-eclampsia/HELLP Disseminated intravascular coagulation

Introduction

Haemophilia is a group of congenital disorders of blood coagulation that arise as a result of a deficiency of clotting factor proteins, which are essential to the normal intrinsic coagulation pathway. The classic form, haemophilia A, is attributable to deficiency of Factor VIII while haemophilia B (also known as Christmas disease) is attributable to deficiency of Factor IX. Both these diseases have a classic X-linked pattern of inheritance and thus affect males while female carriers may also have mild deficiency of the appropriate coagulation factor.

Haemophilia A is the commoner disease (80%), with an incidence of 1 in 8000–10 000 live male births, compared to an incidence of 1 in 25 000–30 000 for haemophilia B (20%).

Pathophysiology

The normal clotting system is activated in the presence of vascular injury to produce: (1) vascular spasm, (2) platelet plug formation, (3) coagulation: factor activation and the production of fibrin. Normal coagulation of blood is dependent on the generation of adequate thrombin via the clotting cascade. Deficiency of Factor VIII or IX reduces the amplification of the clotting cascade and causes haemophilia. The severity of the bleeding disorder is inversely related to the level of functional factor present and is categorized into mild, moderate and severe disease:

Clinical features

Haemophilia A and B are clinically indistinguishable and symptoms vary according to the severity of the inherited disorder. Mild disease may not present till adulthood, whereas moderate to severe disease usually presents in infancy or early childhood.

Bleeding in haemophilia tends to occur spontaneously or following minor trauma and is typically delayed and persistent. This is because although initial platelet ‘plugging’ function is normal, the subsequent coagulation ‘cascade’ response is abnormal. This delay is usually hours, occasionally days. Once bleeding occurs it may persist for days or even weeks. Patients who are severely affected may present with bleeding episodes on a weekly basis.

The most common manifestations of haemophilia are:

Patients may also present with a complication of therapy. Most haemophiliac patients treated before 1985 have been exposed to pathogenic viruses, of which the most important are hepatitis C, hepatitis B and HIV. Of those who received plasma prior to the mid-1980s, 90% are hepatitis B positive, 85–100% are hepatitis C positive, and 60–90% are HIV positive.

Clinical investigation

Investigations are tailored to the individual presentation. A full blood count, blood film and coagulation profiles are useful in the evaluation of first presentations, or major bleed, but unlikely to be helpful in patients with an established diagnosis. Plain radiography of affected joints and computerized tomography (CT) scanning of the head, chest, abdomen and pelvis may be essential to establish the presence or absence of bleeding complications.

The prothrombin time measures primarily Factors II, VII, V and X, thus patients with both haemophilia A and B have a normal prothrombin time, and a normal thrombin clotting time. The partial thromboplastin time measures activation of all factors other than Factor VIII and is prolonged in haemophilia (although it can be normal if factor activity exceeds 30%). Specific factor assays are required to distinguish between haemophilias A and B.

Treatment

Treatment of haemophilia has evolved dramatically in the past 40 years with the discovery in the 1960s that coagulation Factor VIII was concentrated in cryoprecipitate. More recently, highly purified concentrates of Factor VIII and Factor IX have been developed.

Products currently available for the treatment of haemophilia include:

Treatment of acute bleeding episodes primarily involves administration of factor replacement therapy. Complications of bleeding may require specific intervention. Adjunctive therapies include pain relief, rest and immobilization. Specific treatment is influenced by:

Haemophilia patients presenting with suspected bleeds should be triaged as ATS 3 and receive prompt assessment by a senior doctor. For muscle and joint bleeds ‘R.I.C.E.S’ should be initiated on arrival to limit bleeding and reduce pain.

Options for adequate analgesia include:

In the context of pain relief, aspirin (or other platelet-modifying drugs) should be avoided and NSAIDs used with caution. Intramuscular injections should never be administered. In major bleeds there may be a requirement for red cell transfusion. Developing limb compartment syndromes may require surgical decompression and intracranial bleeds may require neurosurgical intervention. In all of these cases, factor replacement must commence as quickly as possible. Management of complex presentations requires a multidisciplinary approach, and early consultation with the relevant state haemophilia centre, especially if the presentation is a major bleed or the patient has inhibitors.

Intravenous cannulation is best performed by a skilled practitioner to help ensure vein preservation. Invasive procedures such as arterial puncture and lumbar puncture must only be performed after clotting factor replacement. Intramuscular injections should be avoided.

Some patients with Factor VIII levels higher than 10% may be successfully treated with 1-amino-8-D-arginine vasopressin (desmopressin, DDAVP). It acts by releasing von Willebrand Factor stored in the lining of the blood vessels. Von Willebrand Factor is a protein that transports Factor VIII in the bloodstream and as such plays an important role in blood clotting. Desmopressin appears to mobilize available Factor VIII stores and may raise Factor VIII activity by a factor of three. If the patient has previously had a documented good response to desmopressin, this can be used as first line therapy for minor bleeding such as haemarthroses.

Desmopressin can be administered intravenously, subcutaneously or by nasal spray. The intravenous dose is 0.3 mcg/kg in 100 mL saline over no less than 45 min. A response should be evident within the first hour. More rapid administration can be associated with blood pressure changes. Side effects, including facial flushing and headache, are usually well tolerated. Tachyphylaxis tends to develop after three or four doses. The antidiuretic properties of desmopressin (which can last up to 24 h after a dose) can produce fluid retention and hyponatraemia (leading to seizures), and the serum sodium levels should be measured before giving further doses. Desmopressin is useful in treating mild and very rarely moderate haemophilia A. It is not of value in severe haemophilia A or with any type of haemophilia B. In serious bleeds or major surgery, desmopressin alone will not control bleeding. In such a case, most patients should also receive Factor VIII concentrate, recombinant Factor VIII replacement and recombinant Factor IX replacement.

Most haemophiliac patients are usually well known to their state haemophilia centres, who often hold specialized treatment protocols for difficult or complex patients. Patients should have their treatment regime cards with them. If not they should be encouraged to do so.

One unit of Factor VIII concentrate provides the amount of Factor VIII activity in 1 mL of normal plasma. Given that a 70-kg adult has a plasma volume of 3500 mL, we can expect that an infusion of 3500 units of Factor VIII will produce 100% Factor VIII activity in a haemophiliac with negligible activity prior to treatment. The half-life of Factor VIII is approximately 12 h. Accordingly, a further dose of 1750 units in 12 hours’ time will again restore 100% activity.

It is not always necessary to provide 100% Factor VIII activity in order to ensure haemostasis: levels of 30–50% may be sufficient in the context of haemarthrosis or dental extraction. Larger infusions should be reserved for life-threatening situations.

Disposition

von Willebrand disease

Factor VIII has an intimate association with von Willebrand factor (vWF). This is an adhesive glycoprotein secreted by endothelium and megakaryocytes, which is required for the normal instigation of platelet plug formation and for stabilization and transport of Factor VIII within the circulation. Thus von Willebrand disease (vWD) is a result of dysfunction, reduction, or a complete lack of the vWF and is often associated with low Factor VIII activity. It is the most common inherited bleeding disorder, affecting 0.1–1% of the population, and affects males and females equally.

Three types of von Willebrand disease are recognized:

Common symptoms of vWD are:

Useful contacts

Introduction

Blood is a living tissue composed of blood cells suspended in plasma; it transports nutrients and oxygen, and facilitates temperature control. An average 70-kg male has a blood volume of about 5 L. The cellular elements comprise red blood cells, white blood cells and platelets, and make up about 45% of the volume of whole blood. Plasma, which is 92% water, makes up the remaining 55%.

Early attempts at blood transfusion were thwarted by adverse reactions. In 1900 Karl Landsteiner demonstrated the ABO blood group system and explained many of the observed severe incompatibility reactions (Table 13.5.1). He won the Nobel prize for medicine in 1930 and went on to discover the Rhesus factor in 1940. The next major advance in transfusion medicine occurred with the development of long-term anticoagulants such as sodium citrate, which allowed extended preservation of blood. Development of refrigeration procedures allowed storage of anticoagulated blood. The addition of a citrate-glucose solution extended the viability of collected blood to several days. The ability to preserve blood for longer than a few hours paved the way for the establishment of the first blood bank in a Leningrad hospital in 1932.

Table 13.5.1 The ABO group system

ABO blood group	Antigens on red cells	Antibody in serum
O	None	Anti-A, Anti-B
A	A	Anti-B
B	B	Anti-A
AB	A, B	None

Transfusion of blood and blood products is now routine and vital to the practice of emergency medicine. As with any prescribed treatment, these products are associated with potential hazards as well as advantages. The hazards are more likely to be encountered with blood products used during emergencies. The blood products available in most Australian emergency departments (EDs) are packed red blood cells, platelets, fresh frozen plasma (FFP), cryoprecipitate, activated Factor VII, prothrombin complex concentrates and other factor concentrates.

In the Australian urban hospital setting, 50% of packed red cells are used for the treatment of anaemia, 22% pre- or perioperatively and 13% for abnormal, excessive or continued bleeding.¹ Medical oncology uses 78% of all platelets.² Forty-one per cent of all FFP is used to correct coagulopathy associated with surgery, 27% to correct coagulopathy in bleeding, 16% to reverse haemostatic disorders in patients having massive blood transfusion, 11.5% for reversal of warfarin effect and the remaining 4.5% for a number of miscellaneous conditions, including liver disease and disseminated intravascular coagulation (DIC).

In the ED setting, blood products are most often administered to patients with acute rather than chronic blood loss. In most EDs, trauma patients are the major consumers of blood products and in this setting they need to be provided rapidly and infused often in large quantities. However, as short stay units are developed, non-time critical transfusions of blood products for other medical indications are increasingly the responsibility of ED staff.

Packed red blood cells

Packed red blood cells are produced from whole blood collections by removing most of the plasma by centrifugation and then resuspending the red cells in citrate-based anticoagulant-preservative solution to prolong storage time. Each unit of packed cells contains approximately 200 mL of red cells. Transfusion of one unit can be expected to raise the haematocrit by 3% and the haemoglobin by 10 g/L provided there is no ongoing blood loss.

Packed red cells are the blood product most commonly prescribed in the ED, the usual indication being the replacement of acute blood loss.³ Transfusion of packed red cells is indicated where the patient’s oxygen-carrying capacity is so impaired that control of bleeding alone, if indeed it can be readily achieved, is regarded as insufficient to take the patient out of danger. Occasionally it may be necessary to transfuse a patient with a primary haematological condition, a failure of erythropoiesis or a haemolytic process. The indication for transfusion is the same as for haemorrhage: a severe reduction in oxygen-carrying capacity. Complex multisystem failure, such as DIC or septic shock, may result in simultaneous blood loss, circulatory collapse, haemolysis and a coagulopathy. Transfusion therapy may be life saving in this context (Table 13.5.2).

Table 13.5.2 Potential indications for red cell transfusion

Haemorrhage

Dilutional anaemia following severe burns

Iron-deficiency anaemia

Megaloblastic anaemia

Anaemia of chronic disorders

Chronic renal failure

Failure of erythropoiesis

Sickle cell disease

Septic shock

Disseminated intravascular coagulopathy

Fluid administration is the cornerstone of management during initial trauma reception and resuscitation, as hypovolaemia is one of the leading causes of death in trauma. Fluid resuscitation must be rapid and appropriate, and should be reassessed and adjusted at regular intervals. This has led to a significant decrease in morbidity and mortality from major trauma.⁴

The American College of Surgeons estimates fluid and blood loss by using six physiological parameters to grade patients into classes I–IV (Table 13.5.3). This allows more accurate characterization of a patient’s immediate haemodynamic status than do laboratory parameters such as haemoglobin and haematocrit. It is recommended that patients in groups III and IV receive blood.

Transfusion is not indicated when alternative haematinic therapy is deemed safe and appropriate. A moderately anaemic patient who is asymptomatic and not bleeding, with some reserve oxygen-carrying capacity, does not require blood transfusion. A haemoglobin of 7 g/dL is sometimes taken as the failsafe point in the decision whether to transfuse, although of course the patient’s unique circumstances need to be taken into account: treat the patient, not the number. The National Health and Medical Research Council together with the Australasian Society of Blood Transfusion have published transfusion guidelines for red blood cells and other products (Table 13.5.4).

Table 13.5.4 Guidelines for transfusion of blood components

Indications	Considerations
Red blood cells
Hb
<70 g/L	Lower thresholds may be acceptable in patients without symptoms and/or where specific therapy is available
70–100 g/L	Likely to be appropriate during surgery associated with major blood loss or if there are signs or symptoms of impaired oxygen transport
>80 g/L	May be appropriate to control anaemia-related symptoms in a patient on a chronic transfusion regimen or during marrow suppressive therapy
>100 g/L	Not likely to be appropriate unless there are specific indications
Platelets
Bone marrow failure	At a platelet count of <10 × 109/L in the absence of risk factors and <20 × 109 in the presence of risk factors (e.g. fever, antibiotics, evidence of systemic haemostatic failure)
Surgery/invasive procedure	To maintain platelet count at >50 × 109/L. For surgical procedures with high risk of bleeding (e.g. ocular or neurosurgery) it may be appropriate to maintain at 100 × 109/L
Platelet function disorders	May be appropriate in inherited or acquired disorders, depending on clinical features and setting. In this situation, platelet count is not a reliable indicator
Bleeding	May be appropriate in any patient in whom thrombocytopenia is considered a major contributory factor
Massive haemorrhage/transfusion	Use should be confined to patients with thrombocytopenia and/or functional abnormalities who have significant bleeding from this cause. May be appropriate when the platelet count is <50 × 109/L (<100 × 109/L in the presence of diffuse microvascular bleeding)
Fresh frozen plasma
Single factor deficiencies	Use specific factors if available
Warfarin effect	In the presence of life-threatening bleeding. Use in addition to vitamin-K-dependent concentrates
Acute DIC	Indicated where there is bleeding and abnormal coagulation. Not indicated for chronic DIC
TTP	Accepted treatment
Coagulation inhibitor deficiencies	May be appropriate in patients undergoing high-risk procedures. Use specific factors if available
Following massive transfusion or cardiac bypass	May be appropriate in the presence of bleeding and abnormal coagulation
Liver disease	May be appropriate in the presence of bleeding and abnormal coagulation
Cryoprecipitate
Fibrinogen deficiency	May be appropriate where there is clinical bleeding, an invasive procedure, trauma or DIC

TTP, idiopathic thrombocytopenia purpura ;DIC, disseminated intravascular coagulation.

Adapted from the National Health and Medical Research Council and Australasian Society Clinical Practice Guidelines on appropriate use of blood components http://www.nhmrc.gov.au/publications/synopses/-files/cp82.pdf

Prior to transfusion, except in an extreme emergency, the patient’s informed consent should be sought, obtained and documented. Rarely, patients may be encountered who refuse transfusion, either on religious grounds or for fear of adverse reactions, particularly with respect to the transfer of viral agents. This can pose problems if the patient appears incompetent to give or withhold informed consent. The situation of an exsanguinating minor whose parents refuse permission to transfuse is particularly difficult. Court orders can be obtained to treat minors without parental consent, and in such an extreme situation many doctors would feel justified in applying to obtain one, even retrospectively.

Precautions when cross-matching and transfusing blood

Although most patients do not require transfusion in the ED, it is often appropriate to ‘group and hold’ or cross-match the patient while in the department. Many hospitals have written protocols detailing the anticipated requirements for a given surgical procedure. Documentation should be meticulous. It is preferable that the person drawing the blood for cross-matching should also fill in and sign the laboratory request form. Most severe incompatibility reactions to blood transfusion result not from exposure to unusual antigens but from an administrative error. Any systematic change in documentation protocols, for example the adoption of an electronic record, needs to be accompanied by an obsessive risk management strategy.

The checking of the compatibility details of blood to be transfused must be meticulous. Blood products should not be left lying around workbenches. Universal precautions must be observed by staff setting up transfusions. Rapid or large transfusions should be via a blood warmer. Blood is transfused intravenously through sterile giving sets containing 170 μm filters. Alternative routes (arterial, intraperitoneal or intraosseous) are only used in exceptional circumstances. Lines for transfusion should be dedicated lines; drugs and other additives should be administered at separate sites. Normal saline is compatible with all blood components.

Pulse, blood pressure and temperature are measured at regular intervals, and particular attention is paid to the patient during the first 25 min of the transfusion. The transfusion is started slowly. The rate at which it continues depends upon the clinical urgency. The usual regime is 500 mL over 1–2 h. As a general rule, the faster the anaemia has developed the more rapidly it needs to be corrected. Rapid infusion techniques may be indicated in patients who appear to be exsanguinating, but over-rapid infusion can precipitate cardiac failure in the elderly. Hypothermia may be a problem if a blood warmer is not used.

Adverse reactions to transfusion

The principal adverse reactions to blood transfusion are listed in Table 13.5.5. Serious adverse reactions are relatively rare (Table 13.5.6) although some are more likely to occur when blood is administered urgently.

Table 13.5.5 Adverse effects of blood transfusion

Immunological transfusion reactions	Transmission of infection
Immediate Febrile non-haemolytic reactions Acute haemolytic transfusion reactions Allergic reactions and anaphylaxis Transfusion-related acute lung injury Delayed Delayed haemolytic transfusion reactions Alloimmunization Transfusion-associated graft versus host disease Hypothermia Dilutional coagulopathy Volume overload

Immunological transfusion reactions

Immunological transfusion reactions may be immediate or delayed in onset.

Transmission of infection

In Australia, blood is tested for ABO and Rh (D) blood groups, red cell antibodies, and the following infections:

In terms of viral safety Australia has one of the safest blood supplies in the world (Table 13.5.7).

Table 13.5.7 Risks of transfusion transmitted infection (per unit tested blood transfused)

Infection	Residual risk
CMV	1 in 127 000
Hepatitis B	Approximately 1 in 660 000
Syphilis	Considerably less than 1 in a million
Hepatitis C	<1 in 10 million
HIV	<1 in 10 million
HTLV I and II	<1 in 10 million
Variant CJD	Possible and cannot be excluded

CMV, cytomegalovirus; HIV, human immunodeficiency virus; HTLV, human T-cell lymphotropic virus; CJD, Creutzfeldt–Jakob disease.

Adapted from Australian Red Cross website: http://www.transfusion.com.au/TRANSFUSIONPOCKETGUIDES/Pocket_RiskConsent.asp (accessed 25th February 2008).

Hypothermia

Red blood cells are stored at 4 °C. Rapid infusion of large volumes of stored blood can contribute to hypothermia. Blood warmers should be used during massive blood transfusion. In addition, other intravenous fluids should be warmed and other measures instituted to maintain patient body temperature (see Ch. 28.2).

Dilutional coagulopathies

Clinically significant depletion of coagulation proteins and platelets is a complication of massive transfusion, secondary to dilution and the consumptive coagulopathy of trauma. Stored red cells are deficient in platelets and clotting factors, and transfusion of large amounts can complicate bleeding when not accompanied by assessment and correction of coagulation disturbances. Coagulation parameters including the prothrobin time (PT), activated partial thromboplastin time (APTT), platelet count and fibrinogen level should be monitored and corrected if deficiencies occur in the presence of abnormal bleeding.

Volume overload

This complication occurs when excessive volume of fluid is administered. Pulmonary oedema is a particular risk in the elderly, in infants and in patients with chronic severe anaemia where the red cell mass is decreased but the blood volume is normal.