Chest Pain in Pregnancy Part3: Pulmonary Embolism

Sign in or subscribe to listen

No me gusta!

The flash player was unable to start. If you have a flash blocker then try unblocking the flash content - it should be visible below.

Nurses Edition Commentary

Kathy Garvin, RN and Lisa Chavez, RN

No me gusta!

The flash player was unable to start. If you have a flash blocker then try unblocking the flash content - it should be visible below.

Scott W., M.D. -

What is the best study to evaluate for more proximal DVT if you have high suspicion and u/s negative?

Rebecca B., MD -

Hi Scott,
Thanks for your question. As with anything, it can be a bit complicated! If you suspect proximal DVT. When doing the compression ultrasound, the ultrasonographer can also directly visualize the iliac being and may see a clot and do a Doppler flow study of the iliac vein (since compression there obviously isn't possible). That alone is fairly sensitive and specific. If the compression ultrasound is positive or the clot is visualized in the iliac, you are done and can initiate anticoagulation. However, if it's negative but you still have high suspicion (for example clinically or the Doppler flow isn't entirely normal), you can go a couple ways with this: (1) start anticoagulation and then do serial ultrasound/Doppler studies over the next 2-7 days or (2) MRI. MRI can be done with time of flight and direct thrombus visualization in a non-contrast MRI. These techniques do not use gadolinium, which is contraindicated in pregnancy. The data on this imaging technique isn't that solid and may depend on the local expertise of your radiologist reading your MRI as it is a bit specialized and not a routine testing technique. Hopefully that helps, it's a bit more nuanced depending on the patient and what is available to you at the time. - R. Bavolek

Paul M. -


I don't get it. For the patient with no YEARS criteria and a D-Dimer over 1000, only 1 in 88 women (1.1%) had a PE. Yet we are supposed to scan all of these pregnant women. I read the paper, and it states that 46% of participants were in the third trimester, and their mean D-Dimer was 1120. So I have to ask, what was the D-Dimer in the 1 of 88 women with a PE? The paper doesn't say, but I would guess that it was well over 1000. My suspicion is that there should b a higher D-Dimer cutoff to scan pregnant women with no YEARS criteria, and this should be studied. In the meantime I think the better take away is that if you don't think a woman has a PE, she most likely doesn't, and the D-Dimer would have to be well over 1000, especially in the third trimester, before I recommend CTPA. Am I wrong?



Rebecca B., MD -

Hi Paul,

Thanks for your comment. I think you ask a fair question and definitely bring up a point for discussion. First, if you don't think that the person has a PE at all, then I would say don't even enter into the door of applying the YEARS criteria. I think of it similarly to a non-pregnant patient who presents with chest pain, shortness of breath, might be tachycardic... but it's clear that they have pneumonia (or COVID but let's not even start with COVID!). For that patient, even though they have "criteria" that would make them high risk by the PERC rule, but I wouldn't even "go there" to apply PERC in the first place because their tachycardia and chest pain are due to infection, not PE. I wouldn't get a D-dimer, I'm just not walking that path.

Just like any clinical decision rule, one would need to apply it only to people that meet their criteria of having the suspected disease in the first place, in this case suspected PE. I think the discussion alongside yours is of this is that some clinicians have an INCREDIBLY low threshold for suspecting/working up PE in pregnant patients and will CT anyone with a very low threshold. If you look at those who entered in the study, they were patients who were pregnant and had suspected PE by "some criteria", history, exam, etc. Of that cohort of 494 with SUSPECTED PE, 252 patients had no YEARS criteria (no signs of DVT, no hemoptysis, PE is NOT the most likely diagnosis), so went into the arm of checking a D-dimer of 1000. Of those 164 had a D-dimer of < 1000 but 11 of them STILL underwent CTPA (all were negative). This tells me that some clinicians are VERY nervous about missing PE in a pregnant patient. So, the point of this study is to figure out a reasonable algorithm and to diminish imaging of these patients. Indeed, looking at the algorithm, CTPA was avoided in 65% of patients in the first trimester (when organogenesis is a big concern), and 32% of patients in the 3rd trimester. Reductions in CTs of these patients is definitely an outcome we want, and indeed was a goal of the authors in creating this algorithm. In addition, I think looking at these rules, the authors have to pick a reasonable "miss rate"... is it 1%? 5% As you state, we don't know what the D-dimer is of the 1 person in the 88 was. Was it 1001? Was it 25000?

Certainly, the modified YEARS algorithm presented here has a ways to go. It's an initial decision rule that was prospectively validated but on a limited number of people. I would suspect that this will be used on several different populations of pregnant patients, re-studied, and calibrated. Or, will it go the way of the San Francisco syncope rule? I'm not sure. But it is the first attempt at an algorithmic based decision rule specifically for pregnant patients with suspected PE and giving some clinicians the peace of mind of NOT imaging a pregnant patient that where they suspect PE.

The last thing I will say that almost always, clinical expertise will be better than a rote clinical decision rule. So, rely on your gestalt if you are comfortable with it. If you don't think the patient has a PE, then you are most likely right, then don't go any further with your investigation.

Thanks, -R. Bavolek

Paul R. -

At the institution I'm employed at, our radiology department has a different algorithm for imaging for PE. One of our radiologists stated to me that due to hyperdynamic cardiac function in late pregnancy, CTA chest often provides inconclusive results due to inaccurate timing of IV contrast administration in pregnant state verses nonpregnant state. Per their protocol, if chest x-ray is negative, they recommend VQ perfusion. If chest x-ray positive for pulmonary abnormality, then proceed with CTA chest.

Rebecca B., MD -

Hi Paul,

Thanks for your comment. Yes, that debate rages on... which is best for PE in pregnancy, CTA or VQ perfusion. I think I've been through 3-4 cycles of this at each place I've worked. As you noted, the hyperdynamic circulation does indeed make CTA in pregnancy not as sensitive of a diagnostic test. In addition, there is a radiation trade off. More breast radiation in CTA and more fetal radiation in the VQ study. I would say that it's also likely that it varies by radiologist. Some places may not have people that are as comfortable reading the VQ and more comfortable reading the CTA despite the contrast timing difficulties. They are both acceptable tests in this patient population, so go with the one that your radiologists prefer because it will be more accurate!

Thanks, -R. Bavolek

Timothy R. W., M.D. -

Better late than never (Posting this in July but I have been traveling a lot and did not have time until now)
After listening to the discussion on the workup of pulmonary embolism in pregnancy I have a few comments:
In the 1990’s we had a “gold standard” test to diagnose pulmonary embolism - pulmonary angiography. We routinely failed to use this test and the vast majority of PE workups were non-diagnostic yet we sent the patient home anyhow. Given the ease of CT angiography we have now swung 180 degrees and seem to test nearly everyone with even the remotest chance of a PE making this diagnosis in many patients that would never have been considered in the past. This high utilization of CTA might be good medicine if the testing and the treatment were cheap and without risk - but neither is the case. I believe neither our 1990’s approach nor our current approach is the best care we can provide to our patients, especially in the pregnant patient where the radiation risks of our testing are still not well defined. I think we need to step back a moment and re-consider the good data we also have from the 1990’s that clearly shows we do NOT need to detect every pulmonary embolism if our goal is a patient oriented outcome and not a test oriented outcome.

Steve Hartsell and I wrote an article on this topic that was published in the Annals of Emergency Medicine 20 years ago. It was titled Pulmonary Embolism: Making sense of the diagnostic evaluation. We argued that the pulmonary angiographic approach - the PIOPED “gold standard” - was too sensitive and was routinely ignored and therefor it was time to adopt an “outcome” standard (which in today’s parlance would be called a patient oriented outcome) where it was ok to miss some pulmonary emboli as long as the patients did well. Interestingly our solution was to use d-dimer, lower extremity duplex and CTA to eliminate formal angiography – which is what is being done now but to a degree where I doubt outcomes are being improved in many of the patients and harm is likely occurring in some. Our patient oriented outcome approach was supported by multiple studies showing that a large percentage of PE diagnosis were being missed, but as long as there was not an embolism in waiting (DVT) and the patient had adequate cardiopulmonary reserve they could tolerate non-diagnosis and did just fine (plus they did not undergo the risk of anticoagulation nor the hassles of pulmonary angiography). (Please refer to our article for a more cogent discussion and the hundred plus supporting references to the argument).
I believe this argument and approach is still valid in 2021, specifically in the pregnant person in whom the diagnostic testing and treatment both have more downsides than in the non-pregnant patient. Rather than subjecting a pregnant woman and her fetus to the radiation risk of CTA (or VQ) and the risk of requiring anticoagulation if a tiny clot load is found in the pulmonary vasculature it might be worth considering an outcome standard as the diagnostic approach. If the patient has good cardiopulmonary reserve and stable vitals and does not have a high pre-test clinical probability of PE but she has a + d-dimer, then assess her lower extremity vasculature for DVT and do an echocardiogram looking for right heart strain. If these tests are both negative – refer her to her OB-GYN with a follow-up duplex within the next week and stop your ED workup. Yes you will miss some PE in these lower probability situations - no doubt - but missing a PE than is hemodynamically minimal in a patient with no other clot waiting to dislodge is likely to be ok based on the studies I am eluding to in our article. In fact some of these emboli may be normal (we all likely have them off and on) given the fact that microscopic thromboemboli reach the lung capillary bed continuously and are rapidly destroyed in the fibrinolytic environment of the pulmonary circulation. Since her hypercoagulable state is self-limited and her cardiopulmonary reserve is adequate she should not require long term anticoagulation. If the clot is that hard to find, perhaps the diagnosis and treatment is worse than the disease.

I admit to not keeping up on this literature since I retired 8 years ago. However, I did find in the past that when these diagnostic options are discussed with the pregnant patient – shared decision making - they were more amendable to this approach than to getting a radiation dose. I am interested in other expert opinions on this topic and approach for the pregnant patient.

Tim Wolfe, MD

Reference: Wolfe, T. R. and S. C. Hartsell (2001). "Pulmonary embolism: making sense of the diagnostic evaluation." Ann Emerg Med 37(5): 504-514.

To join the conversation, you need to subscribe.

Sign up today for full access to all episodes and to join the conversation.

To download files, you need to subscribe.

Sign up today for full access to all episodes.
EM:RAP 2021 February Full episode audio for MD edition 214:23 min - 229 MB - M4AEMRAP 2021 February PA Edition PA Edition 60:19 min - 83 MB - MP3EM:RAP 2021 February German Edition Deutsche 97:11 min - 134 MB - MP3EM:RAP 2021 February Canadian Edition Canadian 16:19 min - 23 MB - MP3EMRAP 2021 February Farsi Edition Farsi 191:42 min - 263 MB - MP3EMRAP 2021 February Spanish Edition Español 43:32 min - 60 MB - MP3EMRAP 2021 February French Edition Français 23:52 min - 33 MB - MP3EMRAP 2021 February Aussie Edition Australian 0:49 min - 1 MB - MP3EM:RAP 2021 02 February Written Summary 604 KB - PDFEM:RAP 2021 February Board Review Answers 152 KB - PDFEM:RAP 2021 February Board Review Questions 109 KB - PDFEM:RAP 2021 February MP3 Files 245 MB - ZIP

To earn CME for this chapter, you need to subscribe.

Sign up today for full access to all episodes and earn CME.

6 AMA PRA Category 1 Credits™ certified by CEME (EM:RAP)

  1. Complete Quiz
  2. Complete Evaluation
  3. Print Certificate