It has been shown that there is a non-linear relationship between the dose of UFH infused and the KCCT. This means that disproportionate adjustments in dose are required depending on the KCCT if under- or over-dosing is to be avoided (Box 23.1). Since the half-life of UFH is 1 h, it would take 5 h (five half-lives of the drug) to reach a steady state. A loading dose is, therefore, administered to reduce the time to achieve adequate anticoagulation. UFH in full dose can also be given by repeated subcutaneous injection, and in these circumstances the calcium salt appears to be less painful than the sodium salt. Opinions differ as to whether the subcutaneous or intravenous route is preferable. The subcutaneous route may take longer to reach effective plasma heparin concentrations but avoids the need for infusion devices.

Box 23.1 Guidelines to control unfractionated heparin (UFH) treatment

Source: Modified from Fennerty et al. (1986) and reproduced in British Committee for Standards in Haematology (1998).

Loading dose

5000 iu over 5 min

Infusion

Start at 1400 iu/h (e.g. 8400 iu in 100 mL of normal saline over 6 h). Check after 6 h. Adjust dose according to ratio of the KCCT to the control value using the values below

KCCT ratio Infusion rate change

>7.0	Discontinue for 30 min to 1 h and reduce by >500 iu/h
>5.0	Reduce by 500 iu/h
4.1–5.0	Reduce by 300 iu/h
3.1–4.0	Reduce by 100 iu/h
2.6–3.0	Reduce by 50 iu/h
1.5–2.5	No change
1.2–1.4	Increase by 200 iu/h
<1.2	Increase by 400 iu/h

After each dose change, wait 10 h before next KCCT estimation unless KCCT >5, when more frequent (e.g. 4-hourly) estimation is advisable. Developed using Diogen (Bell and Alton); local validation may be necessary.

KCCT, Kaolin–cephalin clotting time.

Heparin is normally used in the immediate stages of venous thrombosis and PE until the effects of warfarin become apparent. In the past, it has been continued for 7–10 days, but recent evidence indicates that around 5 days of therapy may be sufficient in many instances. This shorter treatment may also reduce the risk of the rare but potentially very serious complications of severe HIT, which normally occurs after the sixth day. LMWH should be administered for at least 5 days or until the INR has been in the therapeutic range for two successive days, whichever is the longer. They have largely replaced UFH, since they can be given subcutaneously (without a loading dose), and without routine monitoring. A full blood count should be ordered after 5 days on LMWH and throughout the duration of LMWH treatment to monitor for heparin-related thrombocytopenia. Patients with previous exposure to heparin within the past 100 days should also have a platelet count performed before the second dose of heparin is administered (Winter et al., 2005).

Heparinoids

Danaparoid is a heparinoid that is licensed for prophylaxis of deep vein thrombosis in patients undergoing general or orthopaedic surgery. It is a mixture of the low molecular weight sulphated glycosaminoglycuronans: heparin sulphate, dermatan sulphate and a small amount of chondroitin sulphate. It acts by inhibiting factor Xa and, like LMWHs, is given by subcutaneous injection. It normally has a low cross-reactivity rate with heparin-associated antiplatelet antibodies and if this is not present can be used in the treatment of individuals who develop HIT but still need ongoing anticoagulation. It is administered intravenously, with monitoring of anti-Xa activity only required in those at high risk of bleeding, for example renal insufficiency. It should be avoided in severe renal insufficiency and severe hepatic insufficiency.

Hirudins

Lepirudin, a recombinant hirudin, is licensed for anticoagulation in patients with type II (immune) HIT who require parenteral antithrombotic treatment. The dose of lepirudin is adjusted according to the APTT, and it is given intravenously by infusion. Haemorrhage is greater in those with poor renal function. Severe anaphylaxis occurs rarely in association with lepirudin treatment and is more common in previously exposed patients (British Committee for Standards in Haematology, 2006b). Bivaluridin is an analogue of hirudin, but acts as a direct thrombin inhibitor. It is licensed for anticoagulation in patients undergoing percutaneous coronary intervention (PCI). It has to be administered parenterally and the activated clotting time (ACT) is used to assess its activity. Haemorrhage is also an important adverse effect of this agent.

Fondaparinux

Fondiparinux sodium is a synthetic pentasaccharide that binds to antithrombin III, thus inhibiting factor Xa but without effect on factor IIa. Therefore, at doses normally used for treatment, it does not significantly affect coagulation tests and routine monitoring of these is not necessary. It has to be given parenterally. It is used for prophylaxis of VTE in high-risk situations and for treatment of acute deep vein thrombosis and treatment of acute PE, except in haemodynamically unstable patients or patients who require thrombolysis or pulmonary embolectomy. It also has an indication for the treatment of unstable angina or non-ST-segment elevation myocardial infarction (NSTEMI) and for treatment of ST-segment elevation myocardial infarction (STEMI). Haemorrhage is the most important adverse effect.

Oral anticoagulants

Warfarin

Although not the only coumarin anticoagulant available, warfarin is by far the most widely used drug in this group because of its potency, duration of action and more reliable bioavailability. Acenocoumarol (nicoumalone) has a much shorter duration of action and phenindione may be associated with a higher incidence of non-haemorrhagic adverse effects. When given by mouth, warfarin is completely and rapidly absorbed, although food decreases the rate (but not the extent) of absorption. It is extremely highly plasma protein-bound (99%) and, therefore, has a small volume of distribution (7–14 L). It consists of an equal mixture of two enantiomers, (R)- and (S)-warfarins. They have different anticoagulant potencies and routes of metabolism.

Both enantiomers of warfarin act by inducing a functional deficiency of vitamin K and thereby prevent the normal carboxylation of the glutamic acid residues of the amino-terminal ends of clotting factors II, VII, IX and X. This renders the clotting factors unable to cross-link with calcium and thereby bind to phospholipid-containing membranes. Warfarin prevents the reduction of vitamin K epoxide to vitamin K by epoxide reductase. (S)-warfarin appears to be at least five times more potent in this regard than (R)-warfarin. Since warfarin does not have any effect on already carboxylated clotting factors, the delay in onset of the anticoagulant effect of warfarin is dependent on the rate of clearance of the fully carboxylated factors already synthesised. In this regard, the half-life of removal of factor VII is approximately 6 h, that of factor IX 24 h, factor X 36 h and factor II 50 h. Some of the variability in response to warfarin may be related to genetic variations in the gene encoding the vitamin K epoxide reductase multiprotein complex (VKORC1 gene).

The effect of warfarin is monitored using the one-stage prothrombin time, for example the international normalised ratio (INR). This test is sensitive chiefly to factors VII, II and X (and to a lesser extent factor V, which is not a vitamin K-dependent clotting factor). However, factor VII, to which the INR is sensitive, is the most important factor in the extrinsic pathway of clotting. The optimum therapeutic range for the INR differs for different clinical indications since the lowest INR consistent with therapeutic efficacy is the best in reducing the risk of haemorrhage. Examples of therapeutic ranges recommended for certain indications are given in Table 23.1 (British Committee for Standards in Haematology, 1998, 2006c).

Table 23.1 Recommended target INRs for different conditions and grade of recommendation

Indication	Target INR	Grade of recommendation
Pulmonary embolus	2.5	A
Proximal deep vein thrombosis	2.5	A
Calf vein thrombus	2.5	A
Recurrence of venous thromboembolism when no longer on warfarin therapy	2.5	A
Recurrence of venous thromboembolism while on warfarin therapy	3.5	C
Symptomatic inherited thrombophilia	2.5	A
Antiphospholipid syndrome	3.5	A
Non-rheumatic atrial fibrillation	2.5	A
Atrial fibrillation due to rheumatic heart disease, congenital heart disease, thyrotoxicosis	2.5	C
Cardioversion	2.5	B
Mural thrombus	2.5	B
Cardiomyopathy	2.5	C
Mechanical prosthetic heart valve (caged ball or caged disk), aortic or mitral^a	3.5	B
Mechanical prosthetic heart valve (tilting disc or bileaflet), mitral^a	3.0	B
Mechanical prosthetic heart valve (tilting disc), aortic^a	3.0	B
Mechanical prosthetic heart valve (bileaflet), aortic^a	2.0	B
Bioprosthetic valve	2.0 (if anticoagulated)	A
Ischaemic stroke without atrial fibrillation	Not indicated	C
Retinal vessel occlusion	Not indicated	C
Peripheral arterial thrombosis and grafts	Not indicated	A
Arterial grafts	2.5 (if anticoagulated)
Coronary artery thrombosis	2.5 (if anticoagulated)
Coronary artery graft	Not indicated	A
Coronary angioplasty and stents	Not indicated	A

INR, international normalised ratio; A, at least one randomised controlled trial (RCT); B, well-conducted clinical trials but no RCT; C, expert opinion but no studies.

a If the valve type is not known, a target INR of 3.0 is recommended for valves in the aortic position and 3.5 in the mitral position.

Source: British Committee for Standards in Haematology (1998, 2006c).

Warfarin is metabolised by the liver via the cytochrome P450 system. Only very small amounts of the drug appear unchanged in the urine. The average clearance is 4.5 L/day and the half-life ranges from 20 to 60 h (mean 40 h). It, thus, takes approximately 1 week (around five half-lives) for the steady state to be reached after warfarin has been administered. The enantiomers of warfarin are metabolised stereo-specifically, (R)-warfarin being mainly reduced at the acetyl side chain into secondary warfarin alcohols while (S)-warfarin is predominantly metabolised at the coumarin ring to hydroxywarfarin. The clearance of warfarin may be reduced in liver disease as well as during the administration of a variety of drugs known to inhibit either the (S) or (R), or both, enantiomers. These are shown in Table 23.2 which is not exhaustive. The number of possible interactions and the potential severity of their outcome mean that it is essential not to prescribe any medicine concomitantly with warfarin until a thorough check on all possible interactions has been undertaken. The British National Formulary contains comprehensive tables listing possible interactions between warfarin and other medicines.

Table 23.2 Some clinically important drug interactions with warfarin

Interacting drug	Effect of interaction on anticoagulant effect	Probable mechanism(s)
Colestyramine	Reduced anticoagulant effect	Impaired absorption and increased elimination of warfarin. N.B. Long-term treatment may cause impaired vitamin K absorption and enhance anticoagulant effect
Colestipol	Reduced anticoagulant effect
Barbiturates	Reduced anticoagulant effect	Induction of warfarin metabolism
Carbamazepine
Griseofulvin
Phenytoin (see also below)
Primidone
Rifampicin
Rifabutin
St John’s wort
Amiodarone	Increased anticoagulant effect	Inhibition of warfarin metabolism
Azapropazone
Chloramphenicol
Cimetidine
Ciprofloxacin
Clarithromycin
Dextropopoxyphene
Erythromycin
Fluconazole
Fluvastatin
Itraconazole
Ketoconazole
Mefenamic acid
Metronidazole
Miconazole
Nalidixic acid
Norfloxacin
Ofloxacin
Phenylbutazone
Sulfinpyrazone
Sulphonamides (e.g. in co-trimoxazole)
Voriconazole
Zafirlukast
Anabolic steroids	Increased anticoagulant effect	Pharmacodynamic potentiation of anticoagulant effect
Bezafibrate
Danazol
Gemfibrozil
Levothyroxine
Phenytoin (see also above)
Salicylates/aspirin (high dose)
Stanozolol
Tamoxifen
Testosterone
Cranberry juice	Increased anticoagulant effect	Mechanism unknown
NSAIDs (including aspirin at all doses)	Increased risk of bleeding	Additive effects on coagulation and haemostasis
Clopidogrel	Increased risk of bleeding	Additive effects on coagulation and haemostasis
Oral contraceptives, oestrogens and progestogens	Reduced anticoagulant effect	Pharmacodynamic antagonism of anticoagulant effect

Vitamin K (e.g. in some enteral feeds)

Renal function is thought to have little effect on the pharmacokinetics of, or anticoagulant response to, warfarin. Some of the variability in warfarin dose requirement is related to genetic polymorphisms of the cytochrome (CYP2C9) mediating the rate of hepatic metabolism of (S)-warfarin. Individuals with the variant isoform (either heterozygotes or in particular homozygotes) metabolise this more active enantiomer more slowly and so require lower doses.

The major adverse effect of warfarin is haemorrhage, which often occurs at a predisposing abnormality such as an ulcer and a tumour. The risk of bleeding is increased by excessive anticoagulation, although this may not need to be present for severe haemorrhage to occur. Close monitoring of the degree of anticoagulation of warfarin is, therefore, important, and guidelines for reversal of excessive anticoagulation are shown in Table 23.3.

Table 23.3 Recommendations for management of bleeding and excessive anticoagulation in patients receiving warfarin

Cause	Recommendation
3.0<INR<6.0 (target INR 2.5)	1. Reduce warfarin dose or stop
4.0<INR<6.0 (target INR 3.5)	2. Restart warfarin when INR <5.0
6.0<INR<8.0, no bleeding or minor bleeding	1. Stop warfarin
6.0<INR<8.0, no bleeding or minor bleeding	2. Restart when INR <5.0
INR>8.0, no bleeding or minor bleeding	1. Stop warfarin
	2. Restart warfarin when INR <5.0
	3. If other risk factors for bleeding give 0.5–2.5 mg of vitamin K (oral)
Major bleeding	1. Stop warfarin
	2. Give prothrombin complex concentrate 50 units/kg or FFP 15 mL/kg
	3. Give 5 mg of vitamin (oral or i.v.)

INR, international normalised ratio; FFP, fresh frozen plasma.

Source: British Committee for Standards in Haematology (1998).

It is also important to reduce the duration of therapy of the drug to the minimum effective period to reduce the period of risk.

Skin reactions to warfarin may also occur but are rare. The most serious skin reaction is warfarin-induced skin necrosis, which may occur over areas of adipose tissue such as the breasts, buttocks or thighs, especially in women, and which is related to relative deficiency of protein C or S. This is important because these deficiencies result in an increased risk of thrombosis, and therefore warfarin may more often be used in such subjects. Preventing excessive anticoagulation in the initial stages of induction of therapy may reduce the severity of the reaction. A dosing schedule which helps to achieve this is shown in Table 23.4.

Day	INR	Warfarin dose (mg)
First	<1.4	10
Second	<1.8	10
	1.8	1
	>1.8	0.5
Third	<2.0	10
	2.0–2.1	5
	2.2–2.3	4.5
	2.4–2.5	4
	2.6–2.7	3.5
	2.8–2.9	3
	3.0–3.1	2.5
	3.2–3.3	2
	3.4	1.5
	3.5	1
	3.6–4.0	0.5
	>4.0	0 (predicted maintenance dose)
Fourth	<1.4	>8
	1.4	8
	1.5	7.5
	1.6–1.7	7
	1.8	6.5
	1.9	6
	2.0–2.1	5.5
	2.2–2.3	5
	2.4–2.6	4.5
	2.7–3.0	4
	3.1–3.5	3.5
	3.6–4.0	3
	4.1–4.5	Miss out next day’s dose then give 2 mg
	>4.5	Miss out 2 days’ doses then give 1 mg

INR, international normalised ratio.

Source: Modified from Fennerty et al. (1984).

Warfarin may also be teratogenic, producing in some instances a condition called chondrodysplasia punctata. This is associated with ‘punched-out’ lesions at sites of ossification, particularly of the long bones but also of the facial bones, and may be associated with absence of the spleen. Although it has been associated predominantly with warfarin anticoagulation during the first trimester of pregnancy, other abnormalities, including cranial nerve palsies, hydrocephalus and microcephaly, have been reported at later stages of pregnancy if the child is exposed.

Although other coumarin anticoagulants are available, in the vast majority of cases these have not been shown to have any clear benefits over warfarin. They may be used occasionally where a patient does not tolerate warfarin. The necessary duration of anticoagulation in venous thrombosis and pulmonary embolus is still uncertain. On the basis of the available evidence, therapy may be required for approximately 6 months after the first deep vein thrombosis or pulmonary embolus. It may be possible to reduce the duration of therapy in patients who have had a postoperative episode since it is likely that the risk factor has been reversed (unless immobility continues). In patients with a second episode, therapy may be required for even longer and in patients with more than two episodes, life-long treatment may be necessary to reduce the risk of recurrence (British Committee for Standards in Haematology, 1998).

Dabigatran

Dabigatran is an orally active inhibitor of both free and clot-bound thrombin (Wittkowsky, 2010). It has a rapid onset of action and does not require laboratory monitoring. Dabigatran etexilate is a pro-drug which is hydrolysed to active dabigatran in the liver. Since 80% of activated dabigatran is excreted unchanged through the kidneys, it should be avoided in patients with severe renal impairment (creatinine clearance < 30 mL/min) and the dose should be reduced in moderate renal impairment (creatinine clearance 30–50 mL/min). Dabigatran is a substrate for the transport protein p-glycoprotein (p-GP), which facilitates renal elimination of certain drugs. Amiodarone, an inhibitor of p-GP, reduces the clearance of dabigatran and so doses should be reduced in patients who are on concurrent treatment with amiodarone. In patients who are on strong p-GP inhibitors such as verapamil and clarithomycin, dabigatran should be used with caution and it should not be used together with quinidine. Drugs such as rifampicin and St John’s Wort, which are potent p-GP inducers, may potentially reduce its efficacy. Dabigatran can be used for prophylaxis of VTE in adults after total hip replacement or total knee replacement surgery (National Institute for Health and Clinical Excellence, 2008). Haemorrhage is the major adverse effect.

Rivaroxaban

Rivaroxaban is an orally active inhibitor of both the ‘free’ and prothombinase complex-bound forms of activated factor X (Xa) (Wittkowsky, 2010). Two thirds of the dose is metabolised, principally by CYP450 enzymes and the remaining third is excreted unchanged in the urine. Like dabigatran, rivaroxaban also appears to be a p-GP substrate and it should be used with caution when prescribed concomitantly with p-GP inhibitors and potent p-GP inducers. It should also be used with caution in patients with creatinine clearance less than 30 mL/min (severe renal impairment) and is contraindicated in those with creatinine clearance less than 15 mL/min. Several CYP3A4 inhibitors and inducers have been shown to affect its metabolism. Some CYP3A4 inhibitors significantly increase the AUC of rivaroxaban, particularly ketoconazole and other azole-antimycotics such as itraconazole, voriconazole and posaconazole and also HIV protease inhibitors such as ritonavir. Therefore, the use of rivaroxaban is not recommended in patients receiving concomitant systemic treatment with these agents. The CYP3A4 inducer rifampicin (and possibly other inducers of this cytochrome) reduces the AUC for rivaroxaban. It is recommended as an option for prophylaxis of VTE in adults after hip or knee replacement surgery (National Institute for Health and Clinical Excellence, 2009). It also does not require laboratory monitoring. Haemorrhage is the major adverse effect.

Fibrinolytic drugs

Thrombolytic therapy is used in life-threatening acute massive pulmonary embolus. It has been used in deep vein thrombosis, particularly in those patients where a large amount of clot exists and venous valvular damage is likely. However, fibrinolytic drugs are potentially more dangerous than anticoagulant drugs, and evidence is not available in situations other than acute massive embolism to show a sustained benefit from their use.

Streptokinase

Streptokinase was the first agent available in this class. It was produced from streptococci and is a large protein that binds to and activates plasminogen, thus encouraging the breakdown of formed fibrin to fibrinogen degradation products. It also acts on the circulating fibrinogen to produce a degree of systemic anticoagulation. Since it is a large protein molecule, it cannot be administered orally and has to be given by intravenous infusion. The half-life of removal from the body is 30 min. It is cleared chiefly by the reticuloendothelial system in the liver.

Its major adverse effect is to increase the risk of haemorrhage but it may also be antigenic and produce an anaphylactic reaction. It may also cause hypotension during infusion and in some patients, particularly those who have been administered the drug within the previous 12 months, a relative resistance to the drug may occur. Thrombolytic therapy is contraindicated in patients who have had major surgery or with active bleeding sites in the gastro-intestinal or genitourinary tract, those who have a history of stroke, renal or liver disease, and those with hypertension. It should also be avoided during pregnancy and the postpartum period.

Alteplase

Tissue plasminogen activator (rt-PA) or alteplase was developed using recombinant DNA technology. Although this agent is much more expensive than streptokinase, it can be used in those situations where streptokinase may be less effective because of development of antibodies, for example within 1 year of previous streptokinase use or where allergy to streptokinase has previously occurred. Because it produces a lesser degree of systemic anticoagulation (it is more active against plasminogen associated with the clot), immediate use of heparin subsequently is necessary to prevent recurrence of thrombosis. Alteplase is also used for acute ischaemic stroke, where its prompt use may improve outcome in carefully selected individuals (National Institute for Health and Clinical Excellence, 2007). At the time of writing, it is the only thrombolytic licensed for this indication (see arterial thromboembolism).

Reteplase and tenectaplase

Reteplase, and more recently tenecteplase, are also fibrin-specific agents and so heparin is required to prevent rebound thrombosis. They are indicated for the treatment of acute myocardial infarction. In this clinical situation, reteplase is administered as an intravenous bolus, followed by a second bolus 30 min later (double bolus), and tenecteplase is given as a single intravenous bolus. They, therefore, have the advantage of convenience of administration compared with alteplase, and are the preferred option in pre-hospital settings, particularly when administered by paramedics (National Institute for Health and Clinical Excellence, 2002a).

Urokinase

Urokinase, like alteplase and streptokinase, can be used for the treatment of deep vein thrombosis and PE. It is also licensed to restore patency in intravenous catheters and cannulas blocked by fibrin thrombi.

Patient care

The patient on oral anticoagulants should be given full information on what to do in case of problems and what circumstances and drugs to avoid. An anticoagulant card with previous INR values and doses should also be provided. The patient should be told of the colour code for the different strengths of warfarin tablet and advised to carry their treatment card at all times. The likely duration of anticoagulant therapy should be made clear to the patient to avoid unnecessary and potentially dangerous prolongations of treatment. Patients who have received a fibrinolytic agent should also carry a card identifying the drug given and the date of administration.

Arterial thromboembolism

Acute myocardial infarction is the commonest clinical presentation of acute arterial thrombosis. Stroke is commonly caused by atherothromboembolism from the great vessels or embolism arising from the heart (approximately 80% of strokes). These two conditions are discussed elsewhere. Peripheral arterial thrombosis or thromboembolism may also occur, most often in the lower limb. Antiplatelet drugs are often used for prophylaxis, but surgical embolectomy and/or fibrinolytic therapy may be needed for treatment of acute thrombotic or thromboembolic events to avoid consequent ischaemic damage.

Aetiology

Arterial thromboembolism is normally associated with vascular injury and hypercoagulability. Vascular injury is most often due to atheroma, itself aggravated by smoking, hypertension, hyperlipidaemia or diabetes mellitus. Although the exact mechanism is not clear, it is thought that platelet aggregation may be induced by the sheer stresses caused by stenosis of an atherosclerotic vessel. This thrombotic material may embolise to cause occlusion further downstream. Hypercoagulability is also a risk factor. It may be associated with increased plasma fibrinogen levels and an increase in circulating cellular components, for example polycythaemia or thrombocythaemia. As mentioned earlier, the thrombus formed in the artery contains a much larger proportion of platelets, possibly reflecting the fact that other blood components that are not as readily adherent may be dissipated by the higher flow rates in the arterial circulation. Oestrogens, by the mechanisms described earlier, are likely to increase the risk of arterial as well as venous thrombosis. Hyperlipidaemia may also increase the risk of hypercoagulability as well as enhancing thrombotic risk through its role in the progression of atheroma and vascular injury.

Treatment and prevention

Aspirin

Aspirin (acetylsalicylic acid) is a potent inhibitor of the enzyme cyclo-oxygenase, which catalyses the production of prostaglandins. It reduces the production of pro-aggregatory prostaglandin, thromboxane A₂ in the platelet, an effect that lasts for the life of the platelet.

Aspirin is well absorbed after oral administration. It is rapidly metabolised by esterases in the blood and liver (so that its half-life is only 15–20 min) to salicylic acid and other metabolites that are excreted in the urine. In the doses used in prophylaxis against thromboembolism, aspirin is largely metabolised by the liver but in overdose, urinary excretion of salicylate becomes a limiting factor in drug elimination.

The major adverse effect of aspirin is gastro-intestinal irritation and bleeding. This problem is much more common with higher doses of aspirin (300 mg or more) that were once used in the prevention of arterial thromboembolism but are less common with the doses (e.g. 75 mg) now recommended. There is evidence that concomitant use of ulcer-healing drugs, particularly proton pump inhibitors, can reduce the risk of non-steroidal anti-inflammatory drug (NSAID)-induced peptic ulceration in patients susceptible to the problem, but haemorrhagic risk may not be significantly reduced. There is also little evidence that buffered or enteric-coated preparations of aspirin are safer in this respect. However, the vast majority of patients tolerate low-dose aspirin well, and it is normally given as a single oral dose of soluble aspirin. Aspirin may also, rarely, induce asthma, particularly in patients with co-existing reversible airway obstruction. Other patients have a form of aspirin hypersensitivity that may result in urticaria and/or angioedema. In this situation, there may be cross-reactivity with other NSAIDs.

Haemorrhagic stroke is a rare but a very serious complication of therapy with aspirin (and with other antiplatelet agents). Recent evidence examining risks and benefits of aspirin has resulted in the recommendation that while long-term use of aspirin, in a dose of 75 mg daily, is of benefit for all patients with established cardiovascular disease, use of aspirin in primary prevention, in those with or without diabetes, is of unproven overall benefit. It must not be given to children or young people under 16 years of age because of the risk of the rare but life-threatening possibility of Reye’s Syndrome (which may cause liver and renal failure).

Clopidogrel

Clopidogrel is a pro-drug that is metabolised in part to an active thiol derivative. The latter inhibits platelet aggregation by rapidly and irreversibly inhibiting the binding of adenosine diphosphate (ADP) to its platelet receptor, thus preventing the ADP-mediated activation of the glycoprotein IIb/IIIa receptor for the life of the platelet. It is an orally active pro-drug and is given once daily for the reduction of atherosclerotic events in those with pre-existing atherosclerotic disease. In this respect, it may be a useful alternative to aspirin in aspirin-allergic subjects but haemorrhage occurs with the same frequency as aspirin, and thrombocytopenia (sometimes severe) may be commoner than with aspirin therapy. Activation to its active metabolite may be subject to a genetic polymorphism of CYP450 2C19 and may also be reduced by the proton pump inhibitor, omeprazole or esomeprazole; so use of alternative gastroprotective agents may need to be considered if required.

Clopidogrel is also licensed for combination use with low-dose aspirin in the management of acute coronary syndrome without ST-segment elevation, when it is given for up to 12 months after the initial event (National Institute for Health and Clinical Excellence, 2004, 2005). Most benefit is obtained in the first 3 months and there is no evidence of benefit of clopidogrel after 12 months in this indication. In combination with low-dose aspirin, clopidogrel is also licensed for acute myocardial infarction with ST-segment elevation. It is recommended for at least 4 weeks in this indication, but the optimum treatment duration has not been established. Finally clopdogrel is sometimes used (with aspirin) in stenting procedures and this sometimes results in long-term use.