Pulmonary emboli almost always originate from venous thrombi, most commonly in the deep veins of the leg. Proximal lower limb DVTs are significantly more likely to embolise than distal (below knee) DVTs. Less commonly, thrombi may form in the iliac veins, veins of the upper limb or within the right heart.

The D-dimer has high sensitivity (97-100%) and is very likely to be positive in the presence of PE. However, the test has poor specificity and positive predictive value as many other conditions can cause a positive result. The d-dimer is useful only in patients with a low pre-test likelihood of PE (as assessed using a tool such as the Well's Criteria) - a negative test in low risk patients makes a diagnosis of PE very unlikely. Ordering this test in non-low risk patients is likely to cloud the clinical picture as a positive result is difficult to interpret.

An ECG should be performed in all patients suspected of pulmonary embolus, mainly to exclude cardiac ischaemia. The most common ECG finding in the case of PE is sinus tachycardia, though right bundle branch block and/or right axis deviation may be present. The classic S1Q3T3 sign (deep S wave in I, pathologic Q wave in 3 and T wave inversion in 3) is both uncommon and non-specific.

An arterial blood gas may reveal a reduced PaO₂, though in small pulmonary emboli the patient may not be hypoxic. PaCO₂ may be normal or reduced due to hyperventilation.

The chest x-ray in PE is generally normal, and is useful for excluding other differential diagnoses such as pneumonia, pulmonary oedema, pneumothorax or pleural effusion. Pulmonary arterial enlargement may be visible in massive PE.

CT pulmonary angiography (CTPA) involves injection of IV contrast followed by imaging of the pulmonary vasculature. If a PE is present, a filling defect may be seen in the main, lobar or segmental arteries; subsegmental PEs are often difficult to visualise. CTPAs are also useful for visualising differential diagnoses for PE such as pneumonia, pulmonary oedema, pneumothorax and pleural effusions. The main drawback of CTPA is that it is relatively contraindicated in patients with renal dysfunction (generally CrCl <30), and a renal physician should be consulted prior to injecting contrast media in these patients. Additionally, CTPA should be avoided in pregnant patients due to the risks of radiation exposure.

The ventilation/perfusion (V/Q) scan is a nuclear imaging modality used to assess for mismatch between ventilation and perfusion using technitium-99 labelled particles; there is almost always a V/Q defect with PE and therefore a negative scan has a high negative predictive value for PE. Other conditions causing a perfusion defect (such as atelectasis, consolidation, ILD, COPD or tumours) will also result in a positive test, and therefore V/Q scans should be avoided in these patients. It is important to note that as an indirect means of assessing for PE, V/Q scans only provide a probability of PE (low, intermediate or high) and cannot confirm the diagnosis in every case.

If PE is confirmed, a venous duplex ultrasound should be performed to assess for a DVT. A DVT is found in 30-50% of cases of pulmonary embolism.

In patients with a massive pulmonary embolus, an echocardiogram may be useful to assess for right ventricular dilatation, strain and ventricular wall hypokinesis. Thrombi are rarely seen on transthoracic echo.

The Well's Criteria are used to stratify risk of PE based on risk factors, and estimate pre-test probability of PE. 

Prevention of pulmonary embolism is as per deep venous thrombosis, with chemoprophylaxis and/or mechanical prophylaxis. See the deep venous thrombosis page for more information.

It is important to consult local guidelines (or a haematologist) regarding the management of pulmonary embolism.