Disorders of the blood cover a wide spectrum of illnesses, ranging from some of the most common disorders affecting mankind (anaemias) to relatively rare conditions such as leukaemias and congenital coagulation disorders. Although the latter are uncommon, advances in cellular and molecular biology have had major impacts on their diagnosis, treatment and prognosis. Haematological changes occur as a consequence of diseases affecting any system and give important information in the diagnosis and monitoring of many conditions.

Functional anatomy and physiology

Blood flows throughout the body in the vascular system, and consists of:

• red cells, which transport oxygen from the lungs to the tissues

• white cells, which defend against infection

• platelets, which interact with blood vessels and clotting factors to maintain vascular integrity and prevent bleeding

• plasma, which contains proteins with many functions, including antibodies and coagulation factors.

Haematopoiesis

Haematopoiesis describes the formation of blood cells, an active process that must maintain normal numbers of circulating cells and be able to respond rapidly to increased demands such as bleeding or infection. During development, haematopoiesis occurs in the liver and spleen, and subsequently in red bone marrow in the medullary cavity of all bones. In childhood, red marrow is progressively replaced by fat (yellow marrow), so that, in adults, normal haematopoiesis is restricted to the vertebrae, pelvis, sternum, ribs, clavicles, skull, upper humeri and proximal femora. However, red marrow can expand in response to increased demands for blood cells.

Bone marrow contains a range of immature haematopoietic precursor cells and a storage pool of mature cells for release at times of increased demand. Haematopoietic cells interact closely with surrounding connective tissue stroma, made up of reticular cells, macrophages, fat cells, blood vessels and nerve fibres (Fig. 24.1). In normal marrow, nests of red cell precursors cluster around a central macrophage, which provides iron and also phagocytoses nuclei from red cells prior to their release into the circulation. Megakaryocytes are large cells which produce and release platelets into vascular sinuses. White cell precursors are clustered next to the bone trabeculae; maturing cells migrate into the marrow spaces towards the vascular sinuses. Plasma cells are antibody-secreting mature B cells which normally represent less than 5% of the marrow population and are scattered throughout the intertrabecular spaces.

Fig. 24.1 Structural organisation of normal bone marrow.

Stem cells

All blood cells are derived from pluripotent haematopoietic stem cells. These comprise only 0.01% of the total marrow cells, but they can self-renew (i.e. make more stem cells) or differentiate to produce a hierarchy of lineage-committed stem cells. The resulting primitive progenitor cells cannot be identified morphologically, so they are named according to the types of cell (or colony) they form during cell culture experiments. CFU–GM (colony-forming unit – granulocyte, monocyte) are stem cells that produce granulocytic and monocytic lines, CFU–E produce erythroid cells, and CFU–Meg produce megakaryocytes and ultimately platelets (Fig. 24.2).

Growth factors, produced in bone marrow stromal cells and elsewhere, control the survival, proliferation, differentiation and function of stem cells and their progeny. Some, such as interleukin-3 (IL-3), stem cell factor (SCF) and granulocyte, macrophage–colony-stimulating factor (GM–CSF), act on a wide number of cell types at various stages of differentiation. Others, such as erythropoietin (Epo), granulocyte–colony-stimulating factor (G–CSF) and thrombopoietin (Tpo), are lineage-specific. Many of these growth factors are now synthesised by recombinant DNA technology and used as treatments: for example, Epo to correct renal anaemia and G–CSF to hasten neutrophil recovery after chemotherapy.

The bone marrow also contains stem cells which can differentiate into non-haematological cells, such as nerve, skeletal muscle, cardiac muscle, liver and blood vessel endothelium. This is termed stem-cell plasticity and may have exciting clinical applications in the future (Ch. 3).

Blood cells and their functions

Red cells

Red cell precursors formed in the bone marrow from the erythroid (CFU–E) progenitor cells are called erythroblasts or normoblasts (Fig. 24.3). These divide and acquire haemoglobin, which turns the cytoplasm pink; the nucleus condenses and is extruded from the cell. The first non-nucleated red cell is a reticulocyte, which still contains ribosomal material in the cytoplasm, giving these large cells a faint blue tinge (‘polychromasia’). Reticulocytes lose their ribosomal material and mature over 3 days, during which time they are released into the circulation. Increased numbers of circulating reticulocytes (reticulocytosis) reflect increased erythropoiesis. Proliferation and differentiation of red cell precursors is stimulated by erythropoietin, a polypeptide hormone produced by renal interstitial peritubular cells in response to hypoxia. Failure of erythropoietin production in patients with renal failure (p. 478) causes anaemia, which can be treated with exogenous recombinant erythropoietin.

Normal mature red cells circulate for about 120 days. They are 8 µm biconcave discs lacking a nucleus but filled with haemoglobin, which delivers oxygen to the tissues. In order to pass through the smallest capillaries, the red cell membrane is deformable, with a lipid bilayer to which a ‘skeleton’ of filamentous proteins is attached via special linkage proteins (Fig. 24.4). Inherited abnormalities of any of these proteins result in loss of membrane as cells pass through the spleen, and the formation of abnormally shaped red cells called spherocytes or elliptocytes (see Fig. 24.8D, p. 999). Red cells are exposed to osmotic stress in the pulmonary and renal circulation; in order to maintain homeostasis, the membrane contains ion pumps, which control intracellular levels of sodium, potassium, chloride and bicarbonate. In the absence of mitochondria, the energy for these functions is provided by anaerobic glycolysis and the pentose phosphate pathway in the cytosol. Membrane glycoproteins inserted into the lipid bilayer also form the antigens recognised by blood grouping (see Fig. 24.4). The ABO and Rhesus systems are the most commonly recognised (p. 1012), but over 400 blood group antigens have been described.

Fig. 24.4 Normal structure of red cell membrane.
Red cell membrane flexibility is conferred by attachment of cytoskeletal proteins. Important transmembrane proteins include band 3 (an ion transport channel) and glycophorin (involved in cytoskeletal attachment and gas exchange, and a receptor for *Plasmodium falciparum* in malaria). Antigens on the red blood cell determine an individual’s blood group. There are about 22 blood group systems (groups of carbohydrate or protein antigens controlled by a single gene or by multiple closely linked loci); the most important clinically are the ABO and Rhesus (Rh) systems (p. 1012). The ABO genetic locus has three main allelic forms: A, B and O. The A and B alleles encode glycosyltransferases that introduce N-acetylgalactosamine (open circle) and D-galactose (blue circle), respectively, on to antigenic carbohydrate molecules on the membrane surface. People with the O allele produce an O antigen, which lacks either of these added sugar groups. Rh antigens are transmembrane proteins.

Haemoglobin

Haemoglobin is a protein specially adapted for oxygen transport. It is composed of four globin chains, each surrounding an iron-containing porphyrin pigment molecule termed haem. Globin chains are a combination of two alpha and two non-alpha chains; haemoglobin A (αα/ββ) represents over 90% of adult haemoglobin, whereas haemoglobin F (αα/γγ) is the predominant type in the fetus. Each haem molecule contains a ferrous ion (Fe²⁺), to which oxygen reversibly binds; the affinity for oxygen increases as successive oxygen molecules bind. When oxygen is bound, the beta chains ‘swing’ closer together; they move apart as oxygen is lost. In the ‘open’ deoxygenated state, 2,3 diphosphoglycerate (DPG), a product of red cell metabolism, binds to the haemoglobin molecule and lowers its oxygen affinity. These complex interactions produce the sigmoid shape of the oxygen dissociation curve (Fig. 24.5). The position of this curve depends upon the concentrations of 2,3 DPG, H⁺ ions and CO₂; increased levels shift the curve to the right and cause oxygen to be released more readily, e.g. when red cells reach hypoxic tissues. Haemoglobin F is unable to bind 2,3 DPG and has a left-shifted oxygen dissociation curve, which, together with the low pH of fetal blood, ensures fetal oxygenation.

Genetic mutations affecting the haem-binding pockets of globin chains or the ‘hinge’ interactions between globin chains result in haemoglobinopathies or unstable haemoglobins. Alpha globin chains are produced by two genes on chromosome 16, and beta globin chains by a single gene on chromosome 11; imbalance in the production of globin chains results in the thalassaemias (p. 1034). Defects in haem synthesis cause the porphyrias (p. 458).

Destruction

Red cells at the end of their lifespan of approximately 120 days are phagocytosed by the reticulo-endothelial system. Amino acids from globin chains are recycled and iron is removed from haem for re-use in haemoglobin synthesis. The remnant haem structure is degraded to bilirubin and conjugated with glucuronic acid before being excreted in bile. In the small bowel, bilirubin is converted to stercobilin; most of this is excreted, but a small amount is reabsorbed and excreted by the kidney as urobilinogen. Increased red cell destruction due to haemolysis or ineffective haematopoiesis results in jaundice and increased urinary urobilinogen. Free intravascular haemoglobin is toxic and is normally bound by haptoglobins, which are plasma proteins produced by the liver.

White cells

White cells or leucocytes in the blood consist of granulocytes (neutrophils, eosinophils and basophils), monocytes and lymphocytes (see Fig. 24.12, p. 1004). Granulocytes and monocytes are formed from bone marrow CFU–GM progenitor cells during myelopoiesis. The first recognisable granulocyte in the marrow is the myeloblast, a large cell with a small amount of basophilic cytoplasm and a primitive nucleus with open chromatin and nucleoli. As the cells divide and mature, the nucleus segments and the cytoplasm acquires specific neutrophilic, eosinophilic or basophilic granules (see Fig. 24.3). This takes about 14 days. The cytokines G–CSF, GM–CSF and M–CSF are involved in the production of myeloid cells, and G–CSF can be used clinically to hasten recovery of blood neutrophil counts after chemotherapy.

Myelocytes or metamyelocytes are normally found only in the marrow but may appear in the circulation in infection or toxic states. The appearance of more primitive myeloid precursors in the blood is often associated with the presence of nucleated red cells and is termed a ‘leucoerythroblastic’ picture; this indicates a serious disturbance of marrow function.

Neutrophils

Neutrophils, the most common white blood cells in the blood of adults, are 10–14 µm in diameter, with a multilobular nucleus containing 2–5 segments and granules in their cytoplasm. Their main function is to recognise, ingest and destroy foreign particles and microorganisms (p. 72). A large storage pool of mature neutrophils exists in the bone marrow. Every day, some 10¹¹ neutrophils enter the circulation, where cells may be circulating freely or attached to endothelium in the marginating pool. These two pools are equal in size; factors such as exercise or catecholamines increase the number of cells flowing in the blood. Neutrophils spend 6–10 hours in the circulation before being removed, principally by the spleen. Alternatively, they pass into the tissues and either are consumed in the inflammatory process or undergo apoptotic cell death and phagocytosis by macrophages.

Under normal conditions platelets are discoid, with a diameter of 2–4 µm (Fig. 24.7). The surface membrane invaginates to form a tubular network, the canalicular system, which provides a conduit for the discharge of the granule content following platelet activation. Drugs which inhibit platelet function and thrombosis include aspirin (cyclo-oxygenase inhibitor), clopidogrel (adenosine diphosphate (ADP)-mediated activation inhibitor), dipyridamole (phosphodiesterase inhibitor), and the IIb/IIIa inhibitors abciximab, tirofiban and eptifibatide (which prevent fibrinogen binding; p. 594).

Fig. 24.7 Normal platelet structure.
(5-HT = 5-hydroxytryptamine, serotonin; ADP = adenosine diphosphate; ATP = adenosine triphosphate)

Clotting factors

The coagulation system consists of a cascade of soluble inactive zymogen proteins designated by Roman numerals. When proteolytically cleaved and activated, each is capable of activating one or more components of the cascade. Activated factors are designated by the suffix ‘a’. Some of these reactions require phospholipid and calcium. Coagulation occurs by two pathways: it is initiated by the extrinsic (or tissue factor) pathway and amplified by the ‘intrinsic pathway’ (see Fig. 24.6).

Clotting factors are synthesised by the liver, although factor V is also produced by platelets and endothelial cells. Factors II, VII, IX and X require post-translational carboxylation to allow them to participate in coagulation. The carboxylase enzyme responsible for this in the liver is vitamin K-dependent. Vitamin K is converted to an epoxide in this reaction and must be reduced to its active form by a reductase enzyme. This reductase is inhibited by warfarin, and this is the basis of the anticoagulant effect of coumarins (p. 1019). Congenital (e.g. haemophilia) and acquired (e.g. liver failure) causes of coagulation factor deficiency are associated with bleeding.

Investigation of diseases of the blood

The full blood count

To obtain a full blood count (FBC), anticoagulated blood is processed through automated blood analysers which use a variety of technologies (particle-sizing, radiofrequency and laser instrumentation) to measure the haematological parameters. These include numbers of circulating cells, the proportion of whole blood volume occupied by red cells (the haematocrit, Hct), and the red cell indices which give information about the size of red cells (mean cell volume, MCV) and the amount of haemoglobin present in the red cells (mean cell haemoglobin, MCH). Blood analysers can differentiate types of white blood cell and give automated counts of neutrophils, lymphocytes, monocytes, eosinophils and basophils. It is important to appreciate, however, that a number of conditions can lead to spurious results (Box 24.1). The reference ranges for a number of common haematological parameters in adults are given in Chapter 29.

24.1 Spurious FBC results from autoanalysers

Result	Explanation
Increased haemoglobin	Lipaemia, jaundice, very high white cell count
Reduced haemoglobin	Improper sample mixing, blood taken from vein into which an infusion is flowing
Increased red cell volume (MCV)	Cold agglutinins, non-ketotic hyperosmolarity
Increased white cell count	Nucleated red cells present
Reduced platelet count	Clot in sample, platelet clumping

Blood film examination

Although technical advances in full blood count analysers have resulted in fewer blood samples requiring manual examination, scrutiny of blood components prepared on a microscope slide (the ‘blood film’) can often yield valuable information (Box 24.2 and Fig. 24.8). Analysers cannot identify abnormalities of red cell shape and content (e.g. Howell–Jolly bodies, basophilic stippling, malaria parasites) or fully define abnormal white cells such as blasts.

24.2 How to interpret red cell appearances

Microcytosis (reduced average cell size, MCV < 76 fL) A
• Iron deficiency • Thalassaemia	• Sideroblastic anaemia
Macrocytosis (increased average cell size, MCV > 100 fL) B
• Vitamin B₁₂ or folate deficiency • Liver disease, alcohol • Hypothyroidism	• Drugs (e.g. zidovudine, trimethoprim, phenytoin, methotrexate)
Target cells (central area of haemoglobinisation) C
• Liver disease • Thalassaemia	• Post-splenectomy • Haemoglobin C disease
Spherocytes (dense cells, no area of central pallor) D
• Autoimmune haemolytic anaemia	• Post-splenectomy • Hereditary spherocytosis
Red cell fragments (intravascular haemolysis) E
• Microangiopathic haemolysis, e.g. HUS, TTP	• DIC
Nucleated red blood cells (normoblasts) F
• Marrow infiltration • Severe haemolysis	• Myelofibrosis • Acute haemorrhage
Howell–Jolly bodies (small round nuclear remnants) G
• Hyposplenism	• Dyshaematopoiesis
• Post-splenectomy
Polychromasia (young red cells – reticulocytes present) H
• Haemolysis, acute haemorrhage	• Increased red cell turnover
Basophilic stippling (abnormal ribosomal RNA appears as blue dots) I
• Dyshaematopoiesis	• Lead poisoning

(DIC = disseminated intravascular coagulation; HUS = haemolytic uraemic syndrome; MCV = mean cell volume; TTP = thrombotic thrombocytopenic purpura)

Bone marrow examination

In adults, bone marrow for examination is usually obtained from the posterior iliac crest. After a local anaesthetic, marrow can be sucked out from the medullary space, stained and examined under the microscope (bone marrow aspirate). In addition, a core of bone may be removed (trephine biopsy), fixed and decalcified before sections are cut for staining (Fig. 24.9). A bone marrow aspirate is used to assess the composition and morphology of haematopoietic cells or abnormal infiltrates. Further investigations may be performed, such as cell surface marker analysis (immunophenotyping), chromosome and molecular studies to assess malignant disease, or marrow culture for suspected tuberculosis. A trephine biopsy is superior for assessing marrow cellularity, marrow fibrosis, and infiltration by abnormal cells such as metastatic carcinoma.

Fig. 24.9 Bone marrow aspirate and trephine.
A Trephine biopsy needle. B Macroscopic appearance of a trephine biopsy. C Microscopic appearance of stained section of trephine. D Bone marrow aspirate needle. E Stained macroscopic appearance of marrow aspirate: smear (left) and squash (right). F Microscopic appearance of stained marrow particles and trails of haematopoietic cells.

Investigation of coagulation

Bleeding disorders

In patients with clinical evidence of a bleeding disorder (p. 991), there are recommended screening tests (Box 24.3).

24.3 Coagulation screening tests

Investigation	Reference range*	Situations in which tests may be abnormal
Platelet count	150–400 × 10⁹/L	Thrombocytopenia
Prothrombin time (PT)	9–12 secs	Deficiencies of factors II, V, VII or X
Prothrombin time (PT)	9–12 secs	Severe fibrinogen deficiency
Activated partial thromboplastin time (APTT)	26–36 secs	Deficiencies of factors II, V, VIII, IX, X, XI, XII
		Severe fibrinogen deficiency
		Unfractionated heparin therapy
		Antibodies against clotting factors
		Lupus anticoagulant
Fibrinogen concentration	1.5–4.0 g/L	Hypofibrinogenaemia, e.g. liver failure, DIC

N.B. International normalised ratio (INR) is used only to monitor coumarin therapy and is not a coagulation screening test.

(DIC = disseminated intravascular coagulation)

^*Ranges are approximate and may vary between laboratories.

Coagulation tests measure the time to clot formation in vitro in a plasma sample after the clotting process is initiated by activators and calcium. The result of the test sample is compared with normal controls. The tissue factor (‘extrinsic’) pathway (see Fig 24.6) is assessed by the prothrombin time (PT), and the ‘intrinsic’ pathway by the activated partial thromboplastin time (APTT), sometimes known as the partial thromboplastin time with kaolin (PTTK). Coagulation is delayed by deficiencies of coagulation factors and by the presence of inhibitors of coagulation, such as heparin. The approximate reference ranges and causes of abnormalities are shown in Box 24.3. If both the PT and APTT are prolonged, this indicates either deficiency or inhibition of the final common pathway (which includes factors X, V, prothrombin and fibrinogen) or global coagulation factor deficiency involving more than one factor, as occurs in disseminated intravascular coagulation (DIC, pp. 201 and 1055). Further specific tests may be performed based on interpretation of the clinical scenario and results of these screening tests. A mixing test with normal plasma allows differentiation between a coagulation factor deficiency (the prolonged time corrects) and the presence of an inhibitor of coagulation (the prolonged time does not correct); the latter may be chemical (heparins) or an antibody (most often a lupus anticoagulant but occasionally a specific inhibitor of one of the coagulation factors, typically factor VIII). Von Willebrand disease may present with a normal APTT; further investigation of suspected cases is detailed on page 1053.

Platelet function has historically been assessed by the bleeding time, measured as the time to stop bleeding after a standardised incision. However, most centres have abandoned the use of this test. Platelet function can be assessed in vitro by measuring aggregation in response to various agonists, such as adrenaline (epinephrine), collagen, thrombin and ADP, or by measuring the constituents of the intracellular granules, e.g. adenosine triphosphate (ATP)/ADP.

Coagulation screening tests are also performed in patients with suspected DIC, when clotting factors and platelets are consumed, resulting in thrombocytopenia and prolonged PT and APTT. In addition, there is evidence of active coagulation with consumption of fibrinogen and generation of fibrin degradation products (D-dimers). Note, however, that fibrinogen is an acute phase protein which may also be elevated in inflammatory disease (p. 82).

Monitoring anticoagulant therapy

The international normalised ratio (INR) is validated only to assess the therapeutic effect of coumarin anticoagulants, including warfarin. INR is the ratio of the patient’s PT to that of a normal control, raised to the power of the international sensitivity index of the thromboplastin used in the test (ISI, derived by comparison with an international reference standard material).

Monitoring of heparin therapy is, on the whole, only required with unfractionated heparins. Therapeutic anticoagulation prolongs the APTT relative to a control sample by a ratio of approximately 1.5–2.5. Low molecular weight heparins have such a predictable dose response that monitoring of the anticoagulant effect is not required, except in patients with renal impairment (glomerular filtration rate less than 30 mL/min). When monitoring is indicated, an anti-Xa activity assay rather than APTT should be used.

Thrombotic disorders

Measurement of plasma levels of D-dimers derived from fibrin degradation is useful in excluding the diagnosis of active venous thrombosis in some patients (see Fig. 24.15, p. 1010).

A variety of tests exist which may help to explain an underlying propensity to thrombosis, especially venous thromboembolism (thrombophilia) (Box 24.4). Examples of possible indications for testing are given in Box 24.5. In most patients, the results do not affect clinical management (p. 1054) but they may influence the duration of anticoagulation (e.g. antiphospholipid antibodies, p. 1055), justify family screening in inherited thrombophilias (p. 1054), or suggest additional management strategies to reduce thrombosis risk (e.g. in myeloproliferative disease and paroxysmal nocturnal haemoglobinuria; p. 1031). Anticoagulants can interfere with some of these assays; for example, warfarin reduces protein C and S levels and affects measurement of lupus anticoagulant, while heparin interferes with antithrombin and lupus anticoagulant assays. Therefore these tests, when required, should be performed when the patient is not taking anticoagulants.