Diagnosing venous thromboembolism

One of the most challenging aspects is knowing which patients to investigate, explains Dr Kerstin Hogg.

Coloured angiogram: DVT causing blood to pool beneath it (Photograph: SPL)
Coloured angiogram: DVT causing blood to pool beneath it (Photograph: SPL)

Recent advances in research and technology have transformed the way DVT and pulmonary embolism (PE) are diagnosed.

Despite more patients being investigated for venous thromboembolism (VTE) and diagnosed with the condition, the total number of deaths from PE remains unchanged.1

In 2008, PE was the certified cause of death in 3,046 adults in England and Wales. This accounts for 0.6 per cent of all deaths and represents only those patients recognised to have the disease.

Seventy per cent of cases diagnosed at postmortem are not suspected during life,2 therefore this figure is likely to significantly underestimate the proportion of deaths attributable to VTE.

One of the most challenging aspects of diagnosing VTE is deciding which patients to investigate. Symptoms from PE are wide ranging and frequently mimic other disease processes.

In addition, VTE tends to occur in patients with other pathology such as trauma, recent abdominal, obstetric or gynaecological operations, recent joint replacement, recent hospital admissions, stroke, lung disease and other illnesses. This makes it harder to spot new symptoms of PE or DVT.

PE has been misdiagnosed as asthma, pneumonia and MI, with the correct diagnosis eventually made at autopsy.

Spot the symptoms
Most PE patients present with isolated unexplained breathlessness. Patients with massive PE experience dizziness, angina-type chest pain and may collapse.

A minority of PE patients present with pleuritic chest pain and only some of these have haemoptysis. In fact, only 5 per cent of patients presenting to an emergency department with pleuritic chest pain have PE.

DVT often causes thigh or calf pain. Limb swelling and plethora are found in patients with occlusive, proximal femoral vein thromboses, but are frequently absent with smaller clots.

It is important to ask about symptoms of PE in people with symptoms of DVT. We now acknowledge the symptoms can be subtle and varied, so the threshold for investigating for VTE has been lowered.

In the past, all patients with suspected VTE were referred to hospital. Today the diagnostic services are frequently found in the community.

When the decision to investigate for PE or DVT is made, the first step is to estimate how likely it is that the patient has the disease. This can be done empirically, or using an evidence-based scoring system.

Research demonstrates that doctors tend to overestimate the likelihood of VTE, so the Wells score may be an appropriate method (see box).3

Clinical feature Score
Cancer (treatment ongoing, within six months or palliative) 1
Paralysis, paresis, or recent plaster immobilisation of a leg 1
Recently bedridden >three days; major surgery <12 weeks="" ago="" br=""> 1
Localised tenderness along the distribution of the deep venous system 1
Entire leg swelling 1
Calf swelling >3cm compared with asymptomatic leg 1
Pitting oedema (greater in the symptomatic leg) 1
Dilated superficial veins (non-varicose) 1
Previously documented DVT 1
Another diagnosis more likely than DVT -2
Score < 3 means DVT is 'unlikely'; 3 or more signifies DVT is 'likely'.

D-dimer test
Estimating clinical probability allows the clinician to identify which patients are suitable for exclusion of VTE using a D-dimer test. D-dimer is a useful rule-out test in patients who have a low probability of disease.

Any patient with an unlikely Wells score for DVT or a low Wells probability of PE can have a D-dimer test. If D-dimer is within the assay's normal range, VTE can be excluded.

Ideally, the D-dimer result would be reviewed the following day. Since D-dimer is performed on the low risk groups, it is reasonable not to anticoagulate the patient on initial presentation.

A useful alternative is to use a point of care test, which gives a result in a few minutes.4

Any patient with an increased clinical probability or an abnormal D-dimer result requires diagnostic imaging according to local arrangements. They should be commenced on low-molecular weight heparin therapy.

The Wells score for DVT does not include IV drug abuse, and neither score was derived on pregnant women. It is advisable to arrange imaging for all pregnant women and IV drug users with possible DVT or PE. Patients who are tachycardic, tachypnoeic, hypoxic or who have a low BP, should be referred directly to hospital without probability assessment.

Common diagnostic mistakes include investigating a patient for VTE because of an elevated random D-dimer (levels vary in normal healthy individuals) and not testing for D-dimer because the patient is elderly, or had a recent operation. A normal D-dimer in a low-risk patient excludes the disease and avoids invasive diagnostic imaging.

Clinical feature Score
Signs of DVT (objective leg swelling and tenderness) 3
IV drug use 3
PE is the most likely diagnosis 3
Heart rate >100bpm  1.5
Prior PE or DVT diagnosis 1.5
Either bed ridden >three days or surgery, either within the past four weeks 1.5
Cancer (treated actively or palliation within past six months) 1
Haemoptysis 1  1
Low probability = <2 points="" intermediate="" probability="" high="">6 points 

Community-based ultrasound may soon become the norm. Importantly, a normal full leg ultrasound (thigh and calf), excludes DVT, however a normal thigh ultrasound (groin to popliteal fossa) has a false negative rate of 5 per cent. To exclude DVT with thigh ultrasound, the patient must have two serial thigh ultrasounds, a week apart.

Reflect on this article and add notes to your CPD Organiser on MIMS Learning


Question 1. D-dimer can be used to exclude venous thromboembolism in any patient suspected of having the disease.  True or false?

Question 2. Is the Wells score the only way to assess clinical probability of PE?
Question 3.  If a patient has had a normal ultrasound scan of the leg, DVT has been excluded as a cause of their symptoms.  True or False?


CPD IMPACT: earn more credits

1. Burge AJ, Freeman KD, Klapper PJ, Haramati LB. Clin Radiol 2008; 63(4): 381-6.

2. Stein PD, Henry JW. Chest 1995; 108(4): 978-81.

3. Wells PS, Anderson DR, Rodger M, et al. Ann Intern Med 2001; 135(2): 98-107.

4. Geersing GJ, Janssen KJM, Oudega R, et al. BMJ 2009; 339: b2990.

Dr Hogg is thrombosis fellow at The Ottawa Hospital, Canada and was formerly a clinical lecturer at the University of Manchester

Have you registered with us yet?

Register now to enjoy more articles and free email bulletins


Already registered?

Sign in