Guideline 36: Antithrombotic Therapy

Antithrombotic Therapy
Section 2: Venous thromboembolism

2.1 Short term prophylaxis in patients with temporary increased risk

This has been addressed in the SIGN guideline on Prophylaxis of Venous Thromboembolism,⁵ which is currently being revised.

2.2 Diagnosis of acute venous thromboembolism

Acute venous thromboembolism should be suspected in patients with a combination of predisposing factors and suggestive symptoms or signs of either DVT or PE. It should be noted that most patients with proven PE do not have clinically evident DVT.

Predisposing factors:
Recent trauma, surgery, medical illness, or immobilisation from any other cause; especially in the presence of known thrombophilia, previous venous thromboembolism, pregnancy, puerperium, age over 40 years, obesity, varicose veins, oestrogen use, malignancy, heart failure, inflammatory bowel disease, or nephrotic syndrome.⁵

Suggestive symptoms or signs:
DVT: unilateral leg pain, swelling, tenderness, increased temperature, pitting oedema, collateral superficial veins
PE: breathlessness, faintness, collapse, chest pain, haemoptysis, tachycardia, tachypnoea, raised jugular venous pressure, focal signs in chest or on chest x-ray.

Clinical prediction guides may be useful in risk stratification.³⁰ Evidence level III

2.2.1 DIAGNOSTIC IMAGING

Less than 50% of patients with clinically-suspected DVT or PE have the diagnosis confirmed when routine diagnostic imaging is performed.^{31,
32} Evidence level III

Diagnostic imaging should be performed expeditiously (within 24 hours if possible) in patients with suspected DVT (venography, ultrasound) or PE (e.g. ventilation-perfusion lung scanning) to minimise exposure to the risks of inappropriate continued full-dose anticoagulation in those patients in whom venous thromboembolism is not confirmed.

2.2.2 DIAGNOSIS OF DVT

It is safe to withhold anticoagulation therapy in patients with clinically-suspected DVT in whom diagnostic imaging does not confirm the diagnosis.^{30,
31, 33,
34, 35,
36} Consequently, routine diagnostic imaging prevents inappropriate and potentially hazardous exposure of patients who do not have DVT to anticoagulant therapy.^{37,
38} Evidence level III

Figure 1

Possible algorithms for initial management of clinically suspected DVT

[Figure 1]

In all patients with clinically-suspected DVT, the diagnosis should be confirmed or excluded by diagnostic imaging, either:

non-invasive testing by ultrasound (compression or Duplex scanning), followed by contrast venography if negative to detect calf DVT and non-occlusive proximal DVT; or
contrast venography (which detects both calf and proximal DVT); or
serial (repeat after 7 days) non-invasive testing by ultrasound (compression or Duplex scanning) to detect proximal extension of calf DVT.

It should be noted that the pre-test clinical probability of DVT strongly modifies both the positive predictive value and the negative predictive value of all diagnostic tests.^{30,
39} Hence when the findings of ultrasound are at variance with a strong clinical impression, venography should be considered, especially when there is a high clinical probability of disease and a negative non-invasive test result.^{30,
39} Conversely, there is some evidence that a single negative ultrasound test may be sufficient to exclude DVT in patients with a low clinical pre-test probability,³⁰ or a normal fibrin D-dimer test.³³ Evidence level III

A single negative ultrasound may be sufficient to exclude DVT in patients with low clinical pre-test probability and/or a normal fibrin D-dimer assay.

Figure 1 shows possible algorithms for initial management of clinical suspected DVT.

2.2.3 DIAGNOSIS OF PULMONARY EMBOLISM

It is safe to withhold anticoagulant therapy in patients with clinically-suspected PE in whom diagnostic imaging does not confirm the diagnosis of PE or DVT, and who have adequate cardiorespiratory reserve.^{31, 32,
40} Consequently, routine objective testing prevents inappropriate and potentially hazardous exposure of patients who do not have PE or DVT to anticoagulant therapy.³⁷ Evidence level III

In up to 25% of cases, ventilation perfusion lung scans are diagnostic of PE and indicate anticoagulant therapy. In up to 25% of cases, lung scans are normal, and exclude PE. In the remaining 50% of cases, lung scans are non-diagnostic, and should be followed by venography or serial (repeat after 7 days) non-invasive testing by ultrasound and/or by spiral CT angiography or pulmonary angiography (if available).^{31,
32, 40,
41, 42} Evidence level III

In all patients with clinically suspected PE, the diagnosis should be confirmed or excluded by an objective test.

This is usually ventilation-perfusion lung scanning, supplemented by objective testing for DVT.⁴³ Pulmonary angiography,⁴³ spiral CT angiography,^{41, 42} or echocardiography^{41, 42} may also demonstrate PE. Evidence level III

Figure 2 shows one possible algorithm for initial management of clinically suspected PE.

The recent guideline from the British Thoracic Society ⁴² should also be consulted when developing local guidelines. The role of newer imaging techniques, as well as fibrin D-dimer assays,⁴⁴ in diagnosis merits further research.

Figure 2

Possible algorithm for initial management of clinically suspected PE

{Figure 2]

2.3 Heparin in acute treatment of DVT or PE

One randomised controlled trial of intravenous heparin and oral anticoagulants versus no treatment in clinically suspected (but not objectively confirmed) PE⁴⁵ and one randomised controlled trial of intravenous heparin plus oral anticoagulants versus oral anticoagulants alone in objectively-confirmed proximal DVT⁴⁶ showed that heparin reduced the risk of further thromboembolism. Two large randomised controlled trials of heparin treatment of DVT observed that failure to reach the lower limit of the target therapeutic range of the APTT ratio (1.5-2.5) was associated with a 10-15-fold increase in relative risk of recurrent thromboembolic events.³² Descriptive studies have indicated high risks of recurrent thromboembolism and death when heparin was not given to patients with suspected or proven DVT or PE, compared to low risks when heparin was given.^{32,
47, 48}Evidence level Ib supported by evidence levels II and III

Randomised trials have shown equivalent efficacy of low molecular weight heparins to unfractionated heparin.^{20,
49, 50,
51} Evidence level Ib

Outpatients with clinically suspected DVT or PE should be referred to hospital for diagnosis and consideration of initial anticoagulation with heparin, to reduce the risk of further venous thromboembolic events which are often fatal or disabling.

In clinically-suspected DVT or PE, heparin should be commenced (unless strongly contraindicated) until the diagnosis is excluded by diagnostic imaging.

2.3.1 INTRAVENOUS UNFRACTIONATED HEPARIN

This is commenced by an intravenous bolus dose (5,000 IU or 75 IU/kg body weight), followed by maintenance intravenous infusion (1,000-1,500 IU/hour). This is the traditional method of heparin treatment in acute DVT or PE, and remains the treatment of choice in massive PE because of its rapid effect, and because of clinical experience.^{45,
47, 48,
52, 53}

Unfractionated heparin requires monitoring by the APTT. APTT ratios below 1.5 are associated with recurrent thromboembolism and ratios above 2.5 are associated with increased risk of bleeding.³¹ Use of protocols for heparin dose adjustment according to APTT ratio results improves the achievement of therapeutic target ranges.^{31,
52, 53,
54} Each laboratory should standardise its own target range for APTT ratio, which should correspond to an anti-Xa level of 0.35-0.70 u/ml or heparin level of 0.2-0.4 IU/ml by protamine titration.⁵⁴ Evidence level Ib and II

Monitoring of the APTT ratio should be performed expeditiously in patients receiving unfractionated heparin, and heparin doses adjusted according to a local protocol, to achieve the therapeutic target range (usually 1.5-2.5) within 24 hours.

The platelet count should be monitored to detect heparin-induced thrombo-cytopenia (as stipulated by the Committee on Safety of Medicines).

2.3.2 SUBCUTANEOUS UNFRACTIONATED HEPARIN

A meta-analysis of randomised controlled trials has shown that 12-hourly subcutaneous unfractionated heparin is as effective in prevention of recurrent thromboembolism in patients with acute DVT, and is at least as safe, as intravenous unfractionated heparin.⁵⁵ There are no published trials of subcutaneous unfractionated heparin in acute PE. Evidence level Ia

Subcutaneous unfractionated heparin is an effective alternative to intravenous unfractionated heparin for the initial treatment of DVT.

Subcutaneous administration may be advantageous in patients with poor venous access (e.g. intravenous drug users); facilitates mobilisation; and may allow outpatient treatment of some patients with DVT. A conventional intravenous bolus (5,000 IU or 75 IU/kg body weight) should be given, and monitoring of the APTT ratio and dose adjustment performed as for IV heparin, to achieve a therapeutic target range and hence minimise risks of recurrent thromboembolism and of bleeding. Use of a weight-based algorithm allows rapid achievement of effective and safe anticoagulation.⁵⁶ The platelet count should be monitored, to detect heparin-induced thrombocytopenia.

2.3.3 LOW MOLECULAR WEIGHT HEPARINS

A meta-analysis of randomised controlled trials of unfractionated heparin versus low molecular weight (LMW) heparins in treatment of DVT suggests that the latter are similarly effective in reducing extension of DVT and the risk of thromboembolic recurrence; possibly carry a lesser risk of major bleeding; and may be associated with lower mortality.²⁰ Dalteparin, enoxaparin and tinzaparin are licensed for this purpose in the UK. Two trials have also shown equivalence in treatment of PE .^{50,
51} Tinzaparin is licensed for this purpose in the UK. Evidence level Ia

Low molecular weight heparins are effective alternative treatment to unfractionated heparin for DVT and PE.

LMW heparins also have the logistic advantages of efficacy as once-daily subcutaneous injections (facilitating outpatient treatment of DVT in some patients);^{21,
22} and of no requirement for routine daily monitoring of coagulation tests due to their more predictable effect, which is advantageous in patients with poor venous access (e.g. intravenous drug users). However, monitoring of the platelet count is suggested, as for unfractionated heparin, because the incidence of heparin-induced thrombocytopenia appears to be similar.²⁰

2.3.4 BASELINE BLOOD SAMPLES WHEN INITIATING ANTICOAGULANT THERAPY FOR DVT OR PE

A baseline citrated sample-for coagulation screen, and thrombophilia screen if indicated (see table 4); and a baseline blood count including platelet count should be sent to the haematology laboratory prior to starting heparin (which can lower antithrombin levels and platelet count) or oral anticoagulants (which can lower protein C and S levels).

Blood samples should also be sent for baseline urea, electrolytes and liver function tests, to exclude renal or hepatic impairment which are cautions for anticoagulant therapy.

2.4 Oral anticoagulants in acute and maintenance treatment of DVT or PE

One randomised controlled trial of intravenous heparin and oral anticoagulants versus no treatment in clinically suspected PE;⁴⁵ one randomised controlled trial of oral anticoagulants versus no anticoagulants for three months in symptomatic calf DVT;⁵⁷ and three randomised controlled trials of high-intensity oral anticoagulants versus low-dose heparin, adjusted-dose heparin, or lower-intensity oral anticoagulants in proximal vein thrombosis^{58,
59, 60} showed that oral anticoagulants reduce the risk of further thromboembolism. Descriptive studies have reported a high risk of recurrent thromboembolism and death when oral anticoagulants were not given, compared to low risks when oral anticoagulants were given.^{48,
61} Evidence level Ib supported by evidence level III

Following initial heparinisation in patients with DVT or PE, maintenance of anticoagulation with oral anticoagulants is recommended in non-pregnant patients.

In intravenous drug users, the advisability of any continued anticoagulant therapy following hospital discharge should be critically assessed, given the poor compliance with medical care, the bleeding risks of femoral vein and artery self-injection under therapeutic anticoagulation, and the lack of venous access for monitoring of oral anticoagulants and for treatments of oral anticoagulant-induced bleeding.

2.4.1 TARGET INR

Randomised controlled trials^{58,
59, 60,
62, 63,
64} indicate the use of a target INR of 2.5, range 2.0-3.0. This is supported by retrospective descriptive studies which have shown a higher risk of recurrence (in patients without cancer) when INR values are below 1.9,⁶⁵ or lower risk of bleeding at INR 2.0-3.0.^{32,
66, 67} Evidence level Ib supported by evidence level III

The optimal target INR during oral anticoagulant therapy for a first episode of venous thromboembolism is 2.5, range 2.0-3.0.

2.4.2 HEPARIN COVER OF INITIAL ORAL ANTICOAGULANT THERAPY

Evidence from animal studies, pharmacodynamic studies, and clinical reports collectively suggest that heparin is required during the first few days of oral anticoagulant therapy for acute thrombosis, to prevent thrombosis due to a prothrombotic imbalance (earlier reduction in protein C, protein S and factor VII than in factors II, IX and X). Such thrombosis may include microvascular thrombosis causing coumarin-induced skin necrosis, especially in patients with deficiencies of proteins C or S.³²

When oral anticoagulant therapy is initiated for treatment of acute DVT or PE it should be overlapped with heparin therapy for 4-6 days and until the INR is >2.0 on two consecutive days.

2.4.3 TIMING OF ANTICOAGULANT THERAPY

Early institution of oral anticoagulants is as effective and safe as delaying their institution for several days.^{62,
63, 64}

Early institution of oral anticoagulants is recommended in most patients. Evidence level Ib

Particularly when intravenous unfractionated heparin is used, this reduction in the duration of heparin allows earlier mobilisation, earlier discharge from hospital, less infusion-related phlebitis, lower treatment costs, and lower incidence of heparin-induced thrombocytopenia.⁶⁴ It has been suggested (but not proven) that patients with massive iliofemoral DVT may benefit from a longer duration of heparin therapy.³²

2.4.4 DURATION OF ORAL ANTICOAGULANT THERAPY

Six randomised trials have compared shorter durations (4-6 weeks) to longer durations (3-6 months); however differences and defects in their study designs make their interpretation difficult^{32, 66} and at present a minimum of three months duration of treatment is recommended.^{31,
32, 48,
57, 58,
59, 60,
61, 62,
63, 64,
65, 66} Evidence level Ib supported by evidence level III

A retrospective analysis of one randomised trial has suggested that four weeks therapy might be as effective as three months therapy in patients with transient risk factors (e.g. surgery).⁶⁸ However, further trials in this area are needed.⁶⁶

The routine recommended duration of oral anticoagulant therapy following a first episode of DVT or PE is for at least three months.

At three months, patients should be assessed for continuing risk factors (e.g. idiopathic, premature or familial presentation; thrombophilias (see table 4); malignancy; chronic infection; inflammatory bowel disease; nephrotic syndrome; thromboembolic pulmonary hypertension).^{66,
69}

The presence of continuing risk factors suggests consideration of anticoagulation long term, or until such risk factors resolve.

The relative risks of recurrence and of bleeding should be balanced in the individual patient.

2.5 Heparins in maintenance treatment of DVT or PE

Adjusted-dose subcutaneous heparin (target APTT ratio 2.0, range 1.5-2.5)⁵⁹ or subcutaneous low molecular weight heparin²² are effective alternatives to oral anticoagulation, e.g. in pregnancy. ⁷⁰ Evidence level Ib supported by evidence level IV

Adjusted-dose subcutaneous heparin may be considered as an alternative therapy for patients in whom oral anticoagulants are either contraindicated (e.g. pregnancy), inconvenient (e.g. distance from monitoring facilities) or ineffective (e.g. in some cancer patients).

It should be noted that heparin therapy for more than four months carries a high risk of osteopenia and bone fractures. LMW heparins may reduce this risk.²²

2.6 Other antithrombotic therapies in treatment of DVT or PE

2.6.1 MASSIVE PE

In patients with massive PE, cardiorespiratory resuscitation and support, urgent intravenous heparinisation, and urgent confirmation of the diagnosis by echocardiography, spiral CT, pulmonary angiography, or lung scanning are appropriate.

Thrombolytic therapy, percutaneous catheter thrombus fragmentation, or pulmonary embolectomy may each be considered, according to local facilities and expertise.^{28,
31, 32,
42, 66}

(See also figure 2.)

2.6.2 MASSIVE DVT

In patients with massive DVT, leg elevation to reduce oedema, urgent intravenous heparinisation, and urgent confirmation of the diagnosis by venography or ultrasound are appropriate.

Thrombolytic therapy or venous thrombectomy may be considered in patients with threatened venous gangrene. ^{28, 31,
32, 66}

2.6.3 ACUTE NON-MASSIVE DVT

In acute non-massive DVT, leg oedema can be reduced initially by leg elevation, and recurrent oedema minimised by mobilisation with a graduated elastic compression stocking and intermittent leg elevation.

2.6.4 INFERIOR VENA CAVA (IVC) FILTER INSERTION

If therapeutic anticoagulation is contraindicated (e.g. due to high bleeding risk), inferior vena cava (IVC) filter insertion (e.g. with a Greenfield caval filter) should be considered.³²

A recent randomised trial suggests that an initial protective effect may be balanced by a later increased risk of recurrent DVT.⁷¹Temporary filters may reduce this risk, and are under evaluation. Evidence level Ib

Temporary filters may reduce this risk, and are under evaluation.

2.6.5 GRADUATED ELASTIC COMPRESSION STOCKINGS

A randomised controlled trial has shown a reduction in the incidence of severe post-thrombotic leg syndrome from 23% to 11% over two years.⁷² Evidence level Ib

Graduated elastic compression stockings should be worn on the affected leg following proximal DVT for at least two years to reduce the incidence of severe post-thrombotic leg syndrome.

2.7 Recurrent venous thromboembolism

Clinically suspected recurrent venous thromboembolism during oral anticoagulant therapy should be fully investigated, including obtaining objective evidence of recurrence; intensity of oral anticoagulant effect at the time of recurrence (an INR below 2.0 is an indication for heparin therapy, followed by warfarin until target INR is reached); presence of continuing risk factors, especially malignancy and thrombophilias ⁶⁹ (see section 2.10); and hereditary warfarin resistance.⁶⁶

Prophylactic options in documented recurrence despite an oral anticoagulant effect within the therapeutic range include:^{32, 66}

higher-intensity warfarin (target INR 3.5, range 3.0-4.0); or
subcutaneous APTT-monitored unfractionated heparin, or subcutaneous unmonitored LMW heparin (see sections 2.3.2 and 2.3.3); or
insertion of an IVC filter (see section 2.6.4).

2.8 Secondary prophylaxis of DVT

All patients with DVT, and their general practitioners, should be advised of the increased risk of recurrence in future high risk situations (trauma, surgery, medical illness, immobilisation, pregnancy, puerperium) and of the need for prophylaxis.⁵

2.9 Venous thromboembolism in pregnancy

The management of pregnancy-related venous thromboembolism has been fully discussed in a recent national guideline.⁷⁰

2.10 Identification and management of patients with chronic increased risk (thrombophilias)

Screening for congenital or acquired thrombophilias (deficiencies of antithrombin, protein C or protein S; activated protein C resistance which is usually due to the coagulation factor V Leiden mutation; the prothrombin 20210A mutation; lupus anti-coagulants which are usually antiphospholipid antibodies) is recommended in the groups of patients listed in Table 4.⁷³

Patients with such thrombophilias should be referred to consultant haematologists or centres with expertise in management of thrombophilia, which includes:

genetic counselling
general education and advice on the additive risks of obesity and oestrogen-containing oral contraceptives or hormone replacement therapy
prophylaxis: either short term anticoagulant prophylaxis during periods of increased risk of DVT or PE (pregnancy, puerperium, trauma, surgery, or medical illness) or in selected patients long term anticoagulation with warfarin.^{5, 7}

Table 4

Indications for screening for thrombophilias

Venous thromboembolism before the age of 45 years
Recurrent venous thromboembolism or thrombophlebitis
Venous thrombosis in an unusual site (e.g. mesenteric, cerebral)
Unexplained neonatal thrombosis
Skin necrosis, particularly if on coumarins
Arterial thrombosis before the age of 40 years (antiphospholipid syndrome)
Relatives of patients with thrombophilic abnormality
Patients with clear family history of venous thrombosis
Unexplained prolonged activated partial thromboplastin time (APTT) (lupus anticoagulants)
Selected patients with recurrent fetal loss, idiopathic thrombocytopenic purpura, or systemic lupus erythematosus (antiphospholipid syndrome)

(Modified from the British Committee for Standards in Haematology guideline on the investigation and management of thrombophilia ⁷³)

Further randomised trials are required to establish optimum management of thrombophilias.

Recent retrospective or modelling studies suggest that in patients found to have only factor V Leiden (whose prevalence is about 2.5% in the Scottish population⁷⁴) after a first episode of DVT, long term anticoagulation with warfarin may carry a higher relative risk of major bleeding compared to prevention of recurrent DVT.^{75, 76} Evidence level III