Lung Ultrasound - Making a diagnosis


By disease:

Normal
Lung siding - all points
A-lines - all points
Absence of other signs - all points

Pleural effusion
Quad sign - all points but mostly Phrenic and PLAPS points
Sinusoid sign - where the effusion seen

Pneumonia
Can be present in any location so any of:
B-lines
and abolished sliding - any point
Unilateral B-lines (A-lines on the other side) - any point
Anterior alveolar consolidation (high specificity) - upper and lower BLUE points
Lateral or posterior consolidation (present in 90%) - Phrenic and PLAPS points.

Note sliding is abolished when inflammation causes adherences - plural symphysis. This is common in pneumonia and ARDS.

Pulmonary oedema
Bilateral antero-lateral B-lines (upper and lower BLUE and Phrenic points) +/- alveolar consolidation (PLAPS point)
Haemodynamic (cardiogenic)
- B lines start laterally and progress anteriorly with increasing severity.
- In symptomatic patients B-lines will be anterior, widespread and bilateral.
- B-lines quickly disappear after treatment of pulmonary oedema.
Permeability-induced oedema (ALI/ARDS)
- Fluid descends to dependent areas.
- The B-lines are therefore posterior, less widespread and more likely to be asymmetrical.
- Lung sliding may be reduced or absent (as it is an inflammatory process the pleura can get stuck together).
- Anterior subpleural consolidations may be present.
- Spared areas of normal parenchyma may be present.
- Pleural line abnormalities (irregular thickened fragmented pleural line) may be present.

PE
Normal anterior lung US is a very sensitive (though not specific) sign of PE. Peripheral infarcts can sometime be seen with US but these are neither sensitive or specific and are a feature of small rather than major emboli.
The positive predictive value of DVT on US for PE was 89% but this increased to 94% when combined with normal anterior lung US.
Common to have co-existing postero-lateral consolidation (about 50%) - Phrenic and PLAPS points.
Those patients with normal lung US are more likely to have a conclusive test on V/Q scanning so reducing the need for CT irradiation.


COPD / Asthma
Normal or reduced sliding

Pneumothorax
A-lines
No sliding
(96% specificity, near 100% sensitivity)
Lung point
(100% specificity)

Pulmonary fibrosis
Multiple B-lines present in a diffuse and non-homogeneous distribution. The distribution of B-lines correlates with computed tomography (CT) signs of fibrosis.
Pleural line abnormalities (irregular, fragmented pleural line).
Subpleural abnormalities (small echo-poor areas).


By signs:

The 10 signs of lung US
  • The bat sign
  • A-lines
  • Lung sliding
  • Quad sign
  • Sinusoid sign
  • Tissue sign
  • Shred sign
  • B-lines
  • Stratosphere sign
  • Lung point

A-lines
Indicate air so either normal lung surface (COPD, asthma, PE, posterior pneumonia)or pathological (pneumothorax).
Lung sliding
Present:
- Normal lung surface as for A-lines
Absent
- Inflammatory adherences - ARDS, pneumonia
- Pleura separated - pneumothorax
- No lung expansion - one lung intubation, pneumonectomy, v.low tidal volume, apnoea, atelectasis, bulla.

Quad and sinusiod signs - pleural effusion
Tissue and shred signs - alveolar consolidation (oedema, infection, blood) or atelectasis. Also commonly seen in PE. If anterior highly specific for pneumonia. If postero-lateral need anterior A-lines and sliding and absence of thrombosis to specify pneumonia.
Anterior B-lines - interstitial oedema (haemodynamic and some pneumonias).
Stratosphere sign (absent lung sliding)- see above
Lung point - pneumothorax

The BLUE Protocol

‘Bedside Lung Ultrasound in Emergency’

Lichtenstein DA, Mezière GA (2008) Relevance of lung ultrasound in the diagnosis of acute respiratory failure: the BLUE protocol.
CHEST July 2008 vol. 134 no. 1 117-125

A tool for acute respiratory failure which provides a 90.5% accuracy of diagnosis and takes only a few minutes. It should be done as part of the physical examination after the history and before other investigations.

MyBLUE
Adapted from: Lichtenstein DA, Mezière. Relevance of lung ultrasound in the diagnosis of acute respiratory failure: the BLUE protocol. CHEST July 2008 vol. 134 no. 1 117-125

Step 1
Examine the upper and lower BLUE points
Check for anterior lung sliding at these points.
If present this rules out pneumothorax - continue to step 2.
If absent - continue to step 2.


Step 2
Look for anterior A or B-lines.
B-lines
If bilateral the diagnosis is pulmonary oedema.
If unilateral the diagnosis is pneumonia.
Consolidation
The diagnosis is pneumonia.
A-lines
Without sliding:
Look for the lung point.
If present the diagnosis is pneumothorax.
If absent the diagnosis is probably pneumothorax. Consider alternative imaging.
With sliding:
Perform venous US
If DVT present the diagnosis is PE.
If DVT absent continue to step 3.

Or put another way….

With lung sliding:
Bilateral B-lines mean pulmonary oedema.
Unilateral B-lines mean pneumonia.
Consolidation means pneumonia.
Bilateral A-lines are normal and means you need to perform venous US.
Without sliding:
B-lines or consolidation mean pneumonia
A-lines mean pneumothorax likely and should prompt a search for the lung point. If this is not found then alternative imaging should be considered.


Step 3
Examine the PLAPS point.
A normal PLAPS will show sliding and A-lines.
An abnormal PLAPS means consolidation and/or an effusion.
If consolidation is present the diagnosis is pneumonia.
If absent the diagnosis is COPD or asthma.

How good is the BLUE protocol?

Lichtenstein DA, Mezière GA (2008) Relevance of lung ultrasound in the diagnosis of acute respiratory failure: the BLUE protocol. CHEST July 2008 vol. 134 no. 1 117-125

The 90.5% diagnostic accuracy was obtained just by looking at the ultrasound images. Combination of these with history, physical examination and other investigations will improve this to closer to 100%.
For example, the combination of the BLUE protocol with echocardiography would have detected the 7% of patients with pneumonia who had bilateral B-lines mimicking haemodynamic pulmonary oedema.
Specificity for each disease 94% or greater.
Sensitivity for each disease 89% or greater (except for PE - 81%).

Haemodynamic Pulmonary oedema
62 of 64 had bilateral anterior B-lines.
In this condition fluid fills the interlobular septa against gravity so B-lines are anterior, bilateral and widespread. There is no intermediate features between an A and B pattern. There is obviously a threshold reached where A lines suddenly change to B lines. B-lines will increase in number and have a smaller distance between them as oedema worsens 7mm - 3mm.
Alveolar consolidation was common in these patients at the PLAPS point (88%), as would be expected due to the dependent nature of severe oedema filling the alveoli, but it is unnecessary to look for this for the diagnosis.
62 of 64 patients had anterior B-lines.

Pulmonary Embolism
Normal anterior lung US is a very sensitive (though not specific) sign of PE. Peripheral infarcts can sometime be seen with US but these are neither sensitive or specific and are a feature of small rather than major emboli.
The positive predictive value of DVT on US for PE was 89% but this increased to 94% when combined with normal anterior lung US.
PE - 20 of 21 had normal anterior US. 81% had DVT visible.
1 patient had anterior consolidation and 50% had postero-lateral consolidation. Co-existing infection was common.
This illustrates why you should follow the protocol - if you just looked posterolaterally you would make the diagnosis of pneumonia incorrectly.
Those patients with normal lung US are more likely to have a conclusive test on V/Q scanning so reducing the need for CT irradiation.

COPD/asthma
Most SCANS were normal as would be expected in conditions that affect the bronchi.
7 of 49 of COPD had an abnormal US which would be accounted for by coexistant infection or a real diagnosis of pulmonary oedema.

Pneumothorax
All had abolished lung sliding and A-lines.
8 of 9 had a visible lung points.

What about other lung diseases?

Note that the BLUE protocol does not include pleural effusion or chronic interstitial disease / fibrosis as a diagnosis. This is because they are rare causes of acute respiratory failure. They are easily detected by methods outside of the protocol.

Combination of the BLUE protocol with history, examination and other investigations (eg bloods) will improve accuracy to closer to 100%.

What about ventilated patients?

The protocol was designed to assess acute respiratory failure in spontaneously breathing patients.
The BLUE protocol will have some limitations in ICU patients.
Virtually all ventilated patients will have a positive PLAPS (atelectasis, effusion).
So, a post-operative patient with atelectasis would be classified as having pneumonia by the BLUE protocol.
But it is still very useful.
Already shown by Lichtenstein et al in 2004 that in ARDS patients lung US has a higher diagnostic accuracy than CXR for effusion, consolidation and interstitial syndrome. Lichenstein et al (2004) Comparative diagnostic performances of auscultation, chest radiography, and lung ultrasonography in acute respiratory distress syndrome. Anaesthesiology 100:9-15.
A recent study published in 2011 in ICM looked at a general population of medical and surgical ICU patients.
Lung US in critically ill patients: comparison with bedside chest radiography. Xirouchaki et al. Intensive Care Med (2011) 37:1488-1493.
Lung US and CXR were compared to CT as the gold standard.
42 patients prospectively studied (84 lungs).
Lungs were divided into 12 regions as described by Lichtenstein in 2004.
The diagnostic accuracy was:
  • Consolidation - 95%
  • Pneumothorax - 92%
  • Effusion - 100%
  • Interstitial syndrome - 94%
The study showed that BLUE protocol is right in looking at fewer areas - region analysis confirmed that diagnostic accuracy would have been just as good if just the anterior BLUE and PLAPS points were examined (discrepancies with CT were in adjacent regions eg lower lateral and lower posterior). In fact the reason the authors examined more areas was because that was what Lichtenstein advocated in earlier publications before he had simplified his approach based on years of data.

Haemodynamic vs permeability-induced oedema

Haemodynamic
Fluid is forced upwards by hydrostatic pressure.
B-lines will therefore be anterior, widespread and bilateral.
B-lines quickly disappear after treatment of pulmonary oedema.

Permeability-induced oedema
Fluid descends to dependent areas.
The B-lines are therefore posterior, less widespread and more likely to be asymmetrical.
Lung sliding may be reduced or absent (inflammation causes exudative fluid which sticks the pleura together).
Anterior subpleural consolidations may be present.
Spared areas of normal parenchyma may be present.
Pleural line abnormalities (irregular thickened fragmented pleural line) may be present.

Guidance for fluid therapy

The absence of B-lines can reassure that fluid will not be detrimental to gas exchange.
Their presence should, in most circumstances, dissuade you from giving further fluid.
B-lines appear with interstitial oedema (prior to alveolar oedema). It has been proposed that the appearance of B-lines be the stopping point when giving a fluid challenge with B-lines corresponding to the top flat portion of the starling curve.
The absence of bilateral anterior B-lines has a positive predictive value of 97% for a PAOP <18mmHg (negative predictive value of 25% - ie you can have non-cardiogenic pulmonary oedema with a PAOP <18). The PPV is 91% for a PAOP <13.

Weaning

Loss of aerated lung (B-lines, consolidation, atelectasis) suggests weaning will be more difficult.
A recent study presented at the ESICM annual conference showed that assessment of lung aeration by ultrasound prior to, and at the end of, a spontaneous breathing trial both successfully predicted extubation distress.
Ultrasound assessment of lung aeration loss during a successful weaning trial predicts postextubation distress. Soumer et al. Thanks to Tom Clark for sending me the abstract.


2012 Consensus Evidence-based Recommendations (pneumothorax)
Intensive Care Med (2012) 38:577–591

Monitoring lung diseases

B-D4-S1 (strong: level A)
In patients with cardiogenic pulmonary oedema, semi-quantification of disease severity may be obtained by evaluating the number of B-lines as this is directly proportional to the severity of congestion.

B-D4-S2 (strong: level A)
In patients with cardiogenic pulmonary oedema, B-lines should be evaluated because it allows monitoring of response to therapy.

P-D3-S1 (strong: level A)
In patients with increased extravascular lung water, assessment of lung re-aeration can be assessed by demonstrating a change (decrease) in the number of B-lines.

P-D3-S2 (strong: level A)
In the majority of cases of acute lung injury or ARDS, ultrasound quantification of lung re-aeration may be assessed by tracking changes in sonographic findings.
Sonographic findings should include assessment of lung consolidation and B-lines.
RL-D2-S1 (strong: level A)
In critically ill patients with acute lung injury or ARDS, ultrasound changes in lung aeration can be semiquantitatively assessed (at a given location on the chest) using the following four sonographic findings, often in progression:
- Normal pattern - Multiple spaced B-lines
- Coalescent B-lines - Consolidation

RL-D2-S2 (strong: level A)
Lung ultrasound is able to monitor aeration changes and the effects of therapy in a number of acute lung diseases, including the following:
– Acute pulmonary edema – Acute respiratory distress syndrome – Acute lung injury – Community-acquired pneumonia – Ventilator-associated pneumonia
– Recovery from lavage of alveolar proteinosis

RL-D3-S5 (strong: level B)
Serial evaluation of B-lines allows monitoring of pulmonary congestion in patients on hemodialysis, but is of undetermined clinical utility.

P-D2-S4 (no consensus: level C)
The semiquantitative techniques of B-line evaluation (see B-D2-S2 and B-D2-S4) are useful as a prognostic indicator of outcomes or mortality in patients with left-sided heart failure.

RL-D6-S1 (strong: level B)
Semiquantitative B-line assessment is a prognostic indicator of adverse outcomes and mortality in patients with acute decompensated heart failure.