Pulmonary Circulation and Thromboembolism

Pulmonary Thromboembolic Disease

Radiological Features

CXR

• This has a low sensitivity and specificity – however it may exclude other causes (e.g. a pneumothorax)

• The most common signs (without infarction):

  • ▪

    Westermark's sign: localized peripheral oligaemia secondary to an embolus lodging in a peripheral artery ▸ this may be associated with proximal arterial dilatation

  • ▪

    Peripheral airspace opacification: this represents pulmonary haemorrhage

  • ▪

    Linear atelectasis: ischaemic injury to type II pneumocytes leads to surfactant deficiency

  • ▪

    Pleural effusions: these are often small

  • ▪

    Central pulmonary arterial enlargement: this is secondary to chronic repeated embolic disease

• Signs associated with infarction:

  • ▪

    Hampton's hump: a pleurally based, wedge-shaped opacity normally seen within the lateral or posterior costophrenic sulcus ▸ the apex of the triangle points toward the occluded feeding vessel with its base against the pleural surface ▸ it rarely contains air bronchograms ▸ it is a rare and non-specific sign

  • ▪

    Consolidation (which may be multifocal): this predominantly affects the lower lobes ▸ it can be seen from 12 h to several days post embolism

  • ▪

    Cavitation: this follows secondary infection at the infarction site or following a septic embolus

  • ▪

    Haemorrhagic pleural effusions: this is seen in 50% of patients

  • ▪

    Serial CXRs: rapid resolution of any parenchymal changes is associated with a non-infarcting PE – infarction normally heals with scarring and localized pleural thickening

Computed tomography pulmonary angiography (CTPA)

• Technical considerations:

  • ▪

    The imaged area encompasses the central and segmental pulmonary arteries (i.e. from the aortic arch to the inferior pulmonary veins)

  • ▪

    Contrast medium concentrations > 120–250 mg/ml can be associated with significant streak artefacts (particularly adjacent to the SVC)

  • ▪

    Accurate timing of data acquisition (post IV contrast administration) is essential in order to achieve optimum pulmonary arterial opacification

  • ▪

    Contrast density < 300–350HU usually implies a non-diagnostic study

    • –

      Patient-related factors affecting opacification: a SVC thrombosis ▸ persistent intracardiac shunts ▸ pregnancy (the increased cardiac output may result in the majority of the contrast residing beyond the pulmonary arteries when imaging)

• CTPA may reliably detect emboli in up to 4 ^th -order vessels (which are 7 mm in diameter)

  • ▪

    Acute emboli: an intravascular filling defect ▸ it may appear as an expanded unopacified vessel

    • –

      ‘Tram track’ appearance: contrast medium flows around or is adjacent to a clot if the vessel is within the image plane

  • ▪

    Chronic emboli: a crescentic thrombus adherent to the arterial wall (which may also be calcified or show evidence of recanalization) ▸ enlargement of the bronchial vessels (representing the collateral supply) ▸ a prominent mosaic attenuation pattern ▸ right heart strain

  • ▪

    Infarction: a peripheral wedge-shaped region of consolidation (analogous to a Hampton's hump on CXR) ▸ this is only a specific sign if the vessel can be traced to the apex of the wedge

• Can be combined with CT venography from calves to IVC

• CTPA in pregnancy: leg USS first choice ▸ radiation dose to fetus at any stage is negligible with CTPA or V/Q scintigraphy (CTPA protocol should still be adapted to reduced dose and to hypercirculatory state of patient) ▸ check thyroid function (iodine in CT contrast may reduce thyroid function) ▸ breast radiation dose is an issue, but reduced with perfusion scintigraphy (with slightly increased fetal dose)

Perfusion (Q) Scintigraphy

• This assesses the distribution of the pulmonary blood flow ▸ particles microembolize within the lung, providing a map of the pulmonary blood flow (only approximately 2% of the capillaries are occluded) ▸ the effective dose is < 1 mSv

  • ▪

    Technique: it is performed using injected microparticles (10–100 µm) of ^99m Tc microaggregate albumin (MAA) and performed supine to maximize blood flow to the lung apices ▸ the preparation should not be ‘backflushed’ with blood in order to prevent coagulation and ‘hot spots’ forming on the lungs

• A normal perfusion study rules out a PE with almost 100% certainty

Ventilation (V) Scintigraphy

This assesses the distribution of the inhaled air

• Technique: this is performed by the inhalation of krypton-81m, xenon-133, ^99m Tc-DTPA or ‘technegas’ ▸ 8 images are conventionally acquired (anterior, posterior, oblique and lateral projections on both sides)

• ^81m Kr: this is the optimal imaging agent, emitting high-energy photons (190 keV) ▸ its higher photon energy allows the ventilation images to be obtained after the perfusion images as well as allowing matching of both image sets without moving the patient ▸ owing to its short half-life (13 s) it can be continuously administered (including during perfusion imaging) although it means that no washout images are possible

  • ▪

    Disadvantages: there is decreased resolution due to collimator penetration by the high-energy photons ▸ it is expensive

• ¹³³ Xe: although it is cheaper than ^81m Kr, it is a less optimal imaging agent owing to its longer half-life (5.3 days) and low photon energies (80 keV) ▸ ventilation studies need to be performed prior to any perfusion studies (thus preventing Compton scatter from the ^99m Tc into the lower ¹³³ Xe photopeak)

  • ▪

    Single breath inhalation image: a cold spot is abnormal

  • ▪

    Equilibrium phase: tracer activity corresponds to aerated lung

  • ▪

    Washout phase: tracer retention corresponds to areas of air trapping (e.g. COPD)

• ^99m Tc-DTPA and technegas aerosols: these are administered via a nebulizer during inspiration ▸ the aerosol imaging provides a non-physiological static image of lung ventilation with central airway deposition demonstrated ( ^81m Kr allows dynamic imaging) ▸ technegas aerosols and ^81m Kr both provide better images than DTPA

  • ▪

    Disadvantage: it cannot be administered during perfusion (as both aerosols and MAA are labelled with ^99m Tc)

• V/Q ‘mismatches’: a defect on perfusion imaging that appears normal on ventilation imaging and refers to a region of underperfused lung ▸ it suggests the presence of a non-infarct causing pulmonary embolus (despite the interruption of perfusion, the ventilation still remains intact)

  • ▪

    If an infarct develops, a defect may be seen on both perfusion and ventilation studies (i.e. it will become matched as the area will no longer be ventilated) ▸ the ventilation defect is smaller than the perfusion defect as the peri-infarct lung remains ventilated

• Matched ventilation and perfusion defects: these are caused by pulmonary hypoxic vasoconstriction reducing the blood flow to the poorly ventilated lung ▸ it is commonly seen with obstructive airways disease (e.g. asthma, COPD)

• Reversed mismatch: the ventilation defect is more prominent than the perfusion defect ▸ this can be seen with lobar collapse, pneumonic consolidation, a large pleural effusion and obstructive airway causes

• A normal V/Q study: this has a negative predictive value of 100%

• A high probability (>85%) V/Q study: this can be confidently treated as a PE

• An intermediate (15–85%) or low (<15%) probability V/Q study: this requires further imaging if there is a strong clinical suspicion

• Causes of a false-positive results:

  • ▪

    Pulmonary arterial wall causes: vasculitis ▸ infection ▸ irradiation

  • ▪

    Vascular malformations: arterial agenesis ▸ arteriovenous malformations ▸ surgical shunts

  • ▪

    Extrinsic compression of the pulmonary vasculature: hilar adenopathy ▸ tumour

  • ▪

    Pulmonary arterial luminal block: non-thrombotic material ▸ tumour

  • ▪

    A prior pulmonary embolus that has not completely resolved

Pearls

• The terminology of segmental involvement:

  • ▪

    Non-segmental: it does not conform to a lung segment

  • ▪

    Subsegmental : it involves 25–75% of a bronchopulmonary segment

  • ▪

    Segmental: it involves > 75% of a bronchopulmonary segment

Causes of non-segmental defects ^§§

• Pacemaker artefact

• Tumours

• Pleural effusion

• Trauma

• Haemorrhage

• Bullae

• Cardiomegaly

• Hilar adenopathy

• Atelectasis

• Pneumonia

• Aortic ectasia or aneurysm

Ventilation-perfusion terminology ^§§

V/Q matched defect	Both scans abnormal in same area and of equal size
V/Q mismatch	Abnormal perfusion in an area of normal ventilation or a much larger perfusion defect than ventilation abnormality
Triple-match	V/Q matching defects in a region of chest X-ray abnormality where the X-ray abnormality is of the same size or smaller than the radiographic lesion
Segmental defect	Characteristically wedge-shaped and pleural based and conforms to segmental anatomy of the lung. May be caused by occlusion of pulmonary artery branches Large: > 75% of a lung segment Moderate: 25–75% of a lung segment Small: < 25% of a lung segment
Non-segmental defect	Does not conform to segmental anatomy or does not appear wedge-shaped

Modified PIOPED criteria for pulmonary embolus diagnosis

Open full size image

Pulmonary Venous Hypertension (PVH)

Pearls

• Septal lines vs. blood vessels: the latter are not visible in the outer 1 cm of lung, septal lines do not branch uniformly and are seen with greater clarity

• Septal lines can persist in chronic disease (e.g. chronic PVH with fibrosis) or idiopathic interstitial fibrosis, lymphangitis carcinomatosis and pneumoconiosis

• Pre-existing lung disease influences the pattern/distribution of oedema (e.g. extensive emphysema)