Pulmonary Embolism and the Chest X-ray

The diagnostic evaluation and management of pulmonary embolism (PE) is an ever-controversial and ongoing debate in emergency medicine. As emergency physicians (or med students, residents, others), we are constantly bombarded with questions and conflicting opinions on PE:

1)      Is gestalt equal to validated scoring systems like the Wells Criteria (or modified Wells) or the Geneva Score (or modified Geneva)? (see this paper)

2)      Can you “PERC” a patient or are they not “low-gestalt”?

3)      Do we order too many D-dimers?

4)      Do we scan too many chests, causing harm via excessive radiation or exposure to potentially harmful contrast dye?

5)      Do we need to even know about the tiny subsegmental “lung lint” (credit Casey Parker at http://broomedocs.com) or “pulmonary fluff” (See Andy Neil’s http://emergencymedicineireland.com/2013/01/is-pe-really-all-that-bad-again/?)

6)      VQ or CTPA in pregnant patients? (http://radiology.rsna.org/content/262/2/635.long)

The answers to all these questions can be found elsewhere (see links above and the fantastic Corey Slovis/Jeff Kline talk at Free Emergency Medicine Talks https://itunes.apple.com/ca/podcast/free-emergency-medicine-talks/id489694355).

I had a PE patient the other day who got me excited for another reason: the chest x-ray. That’s right, I still get excited when a patients with a PE has interesting chest x-ray findings

Here’s the CXR:

Official read was left basal atelectasis and possible left pleural effusion. Infiltrate cannot be ruled out.

I thought maybe a Hamptom’s Hump (well-defined pleural-based opacity with a convex border – ref here) might be sitting in there too. The CTPA showed multiple left-sided segmental and subsegmental PE’s, areas of infarcted lung in the left base and small pleural effusion.

Now, my patient had a good story for PE – pleuritic sub-scapular pain, a few weeks post caesarean section. But sometimes we have patients with more vague stories and undifferentiated thoracic pain and/or SOB. Can an abnormal finding on a CXR help you think about PE when you hadn’t already? When working up PE, I find the CXR isn’t terribly helpful most of the time. But here is a reminder of what you may see on CXR, as reported in the 1993 PIOPED study.

• In patient’s with PE (angiographically confirmed), only 12% had normal CXR.
• Normal CXR had a poor NPV of 74%.
• Findings that were most common in PE were atelectasis and/or increased opacity in lower lung parenchyma, and pleural effusions.
• For right-sided hemithorax findings, the following signs had these sensitivities/specificities/PPV/NPV comparing PE to non-PE patients (left-sided findings had fairly similar values)
• Westermark's sign - 14% sensitivity, 92% specificity, 38% PPV, 76% NPV
• Hampton's Hump - 22% sensitivity, 82% specificity, 29% PPV, 76% NPV
• Pleural effusion - 36% sensitivity, 70% specificity, 28% PPV, 76% NPV
• Elevated hemidiaphragm - 20% sensitivity, 85% specificity, 30% PPV, 76% NPV
• Normal CXR, pulmonary edema and vascular redistribution were more common in non-PE films.
• Only CXR sign significantly more common in PE was a right-sided Westermark sign (oligemia).

So bottom line from PIOPED: most PE CXR’s do actually have a finding; unfortunately, these findings generally don’t help in differentiating PE from non-PE.

However, I do think it is important to know the CXR findings you may see with a PE. Case in point - I asked a resident in the ED to look at the CXR in the context of the HPI and tell me the diagnosis: "pneumonia?" This is an easy trap to fall into if you're not thinking PE, or even if you are and you see something "unexpected" on the film, as this resident did. Perhaps next time you see an unexplained “atelectasis or lower lung parenchymal opacity with or without pleural effusion” maybe ask yourself, “Is this a PE?” You can’t make the diagnosis if you don’t think about it.

References:

• Asrani A et al. Urgent findings on portable chest radiography: what the radiologist should know – Review. AJR (2011) 196: S45-S61 (a very interesting read) http://www.ncbi.nlm.nih.gov/pubmed/21606239
• Leung et al. American Thoracic Society Documents: An Official American Thoracic Society/Society of Thoracic Radiology Clinical Practice Guideline—Evaluation of Suspected Pulmonary Embolism in Pregnancy. Radiology (2012) 262:635-646 http://radiology.rsna.org/content/262/2/635.long
• Lucassen W. Clinical decision rules for excluding pulmonary embolism: a meta-analysis. Ann Intern Med. (2011) Oct 4;155(7):448-60  http://www.ncbi.nlm.nih.gov/pubmed/21969343
• Westermark N. On the roentgen diagnosis of lung embolism. Acta Radiol (1938) 19:357–372
• Worsley D et al. Chest radiographic findings in patients with acute pulmonary elbolism: observations from the PIOPED study. Radiology (1993) 189:133-136 http://www.ncbi.nlm.nih.gov/pubmed/8372182