iPhoneIcon_BigA study in AJEM sets out to compare diagnostic accuracy between chest x-rays and lung ultrasound for diagnosing pneumonia. Other recent ED studies have consistently shown how lung ultrasound outperforms chest x-rays when diagnosing pneumonia.

The study
The study compares diagnostic accuracy of lung ultrasound, performed by trained ED-ologists, versus chest x-rays in  pneumonia patients.

All patients that were 18 years or older who were suspected to have acute pneumonia were included. Acute pneumonia was suspected if the patients had at least three of temperature ≥ 38℃, cough, dyspnea, heart rate > 100 bpm or saturated <92% on ambient air. Apart from normal standard of care pathway, including chest x-ray, these patients also underwent a lung ultrasound examination performed by ten ED physicians who all underwent a two day training programme in lung ultrasound.

The patient´s chests were scanned in eight sectors including visualising the back side of the lungs with the patient in a sitting position. Ultrasound diagnosis of pneumonia relied on either an ´alveolar´ or an ´interstitial´image.

Alverolar syndrome

Alverolar syndrome

The ´alveolar´ultrasound syndrome is the liver-like appearance of bronchograms whith the pleural line disrupted. It looks much like atelectasis apart from the bronchograms increasing in size with ventilation.

The ´interstitial´ ultrasound syndrome is a lot more meh… It is defined as the appearance of three or more B-lines, comet tails, resonance artefacts in a field of vision.




B-lines are the narrow based rays extending from the pleura all the way to the edge of the screen. B-lines is a  a marker of fluid in the lungs and correlate with extravascular lung water. More comet tails means more water/loss of air.

The gold standard for diagnosing pneumonia would be chest CT. However, exposing all the study subjects to that radiation wouldn’t be justifiable. Instead the final end-point diagnosis, pneumonia or no pneumonia, was performed by a senior clinician who had access to the patients’ complete charts, clinical findings, x-rays, scans and ultrasound imaging.

Screen Shot 2014-05-27 at 19.23.14144 patients fulfilled the inclusion criteriae. 123 of them had pneumonia. Overall 117/123, 95%, were correctly diagnosed using lung ultrasound. Chest x-rays could only identify 60%, 74 out of 123, of the pneumonias.

Screen Shot 2014-05-27 at 19.38.41Lung ultrasound seems even more useful if one looks at patients with a short delay from onset of symptoms to presentation to ED. In patients presenting within 24h of onset of symptoms, lung US found 76% of the pneumonias compared to only 23% with x-rays.

Take home message
A clear win for lung ultrasound. Compares well with several recent studies telling us lung ultrasound outperforms chest x-rays for diagnosing pneumonia or other chest pathology. We could spare our patients from an awful lot of x-rays if we took ultrasound training seriously.

Study lives here:

Bourcier JE, Paquet J, Seinger M, Gallard E, Redonnet JP, Cheddadi F, Garnier D, Bourgeois JM, Geeraerts T.Performance comparison of lung ultrasound and chest x-ray for the diagnosis of pneumonia in the ED. Am J Emerg Med. 2014 Feb;32(2):115-8. doi: 10.1016/j.ajem.2013.10.003. Epub 2013 Oct 9.



This entry was posted in Emergency Medicine, Ultrasound, Uncategorized. Bookmark the permalink.


  1. Pingback: Echographie et pneumonie | thoracotomie

  2. Jakob Mathiszig-Lee says:

    Pretty cool. That said surely at some point we have to say that a doc’s time is better spent seeing patients than performing diagnostic tests.

  3. Jacob Avila says:

    Great study, I especially liked how in the patients that got the CT scan, US was 100% sns, and CXR 52% sns (I think that’s the strongest data of their paper, since CT = gold standard)

    My only criticism is on their definitions. They define alveolar interstitial syndrome as “Disappearance of pleural line associated with aeric or water bronchograms within an image of tissue echogenicity” and interstitial sydnrome as the presence of >3 b-lines.

    This is incorrect./ Several previously published articles describe the following signs:
    1. Interstitial syndrome = >3 b-lines per intercostal space
    2. Alveolar interstitial syndrome = A whole lot of b-lines. (White out of the lung)
    3. Dynamic air bronchograms = Consolidated lung with hyperechoic (white) lines and dots that move with respiration, which are indicative of air trapped in the small airways.
    4. Static air bronchogram = Same thing as dynamic air bronchograms, except that the white dots and lines don’t move with respiration
    5. Fluid bronchogram = airways filled with fluids/secretions following airway obstruction

    In summary: Great article, but they incorrectly labeled air bronchograms (both dynamic and static bronchograms) and fluid bronchograms as “Alveolar interstitial syndrome”. AIS does not show lung parenchyma and is defined as many b-lines that nearly cover the entire US viewing screen.

    Soldati et al. Sonographic interstitial syndrome: the sound of lung water. J Ultrasound Med. 2009 Feb;28(2):163-74.(

    Lichtenstein D, Goldstein I, Mourgeon E, et al. Comparative diagnostic
    performances of auscultation, chest radiography and lung ultrasonography in acute
    respiratory distress syndrome. Anesthesiology 2004;100:9e15. (

    Weinberg The air bronchogram: sonographic demonstration AJR 147:593-595, September 1986 (

    Cortellaro et al Lung ultrasound is an accurate diagnostic tool for the diagnosis of pneumonia in the emergency department Emerg Med J 2012;29:19-23 (

    Reissig et al Lung ultrasound in the diagnosis and follow-up of community-acquired pneumonia. A prospective multicenter diagnostic accuracy study CHEST 2012; 142(4):965–972 (

  4. “Acute pneumonia was suspected if the patients had at least three of temperature ≥ 38℃, cough, dyspnea, heart rate > 100 bpm or saturated <92% on ambient air."

    123/144 had pneumonia. So the inclusion criteria diagnosed pneumonia regardless of CXR or US.

    • Jacob Avila says:

      Great point. 123 out of 144 is 85% correct isn’t that bad. However, other studies don’t agree with that accuracy. A cochrane database article in 2012 stated that “…PNA cannot be reliably diagnosed in children and adolescents with community-acquired pneumonia based on symptoms and signs” (Cochrane Database of Systematic Reviews 2012, Issue 10.) Another study, which included 200 patients with pneumonia found that only 39% had rales 28% had rhonchi. In a study by Van Vugt et al, they found that in 2810 pts with cough got a history and physical. Only 29% had correct diagnosis (CXR gold standard). (Van Vugt Eur Respir J. 2013 Oct;42(4):1076-82 ) Another study by Pisarik found that physical exam findings of pneumonia are present in 4% to 28% of radiographically proven pneumonia (Pisarik The J of Fam Practice 2005 Dec;54(12)).

      So, while being 85% correct in this one study is good, I think the preponderance of evidence shows that the physical exam is unreliable for the diagnosis of PNA. Also, I like my 95% sensitivity better.

Leave a Reply to Charles Bruen Cancel reply

Your email address will not be published. Required fields are marked *