Pulmonary Embolism In Pregnancy


Author: Sarah Sanders, MD (EM Resident Physician, PGY-1, NUEM) // Edited by: Adnan Hussain, MD (EM Resident Physician, PGY-3, NUEM)

Citation: [Peer-Reviewed, Web Publication] Sanders S, Hussain A (2016, January 19). Pulmonary Embolism in Pregnancy. [NUEM Blog. Expert Peer Review by Courtney DM]. Retrieved from http://www.nuemblog.com/blog/pe-in-pregnancy/


Diagnosing a pulmonary embolism (PE) in a pregnant patient is a situation that requires clinicians to employ a high index of suspicion. According to the Centers for Disease Control and Prevention (CDC), PE in pregnancy accounts for 20% of maternal deaths in the United States (1). The presentation is complicated by the fact that symptoms commonly associated with PE in the non-pregnant population can be caused by normal physiologic changes of pregnancy.

A recent study by Gherman et al reviewed the most common presenting symptoms of PE. These included dyspnea (62%), pleuritic chest pain (55%), cough (24%), and sweating (18%) (2). However, Cahill and colleagues analyzed the association of common clinical presentations (chest pain, shortness of breath, O2 <95%, tachycardia, increased A-a gradient, and PaO2 <65mmHg) and the clinical suspicion of pulmonary embolism, and found no statistically significant association of any clinical feature, or grouping of features, that was associated with the confirmed presence of PE (3). Thus, symptomatology is non-specific in the approach of working up a suspected pulmonary embolism in pregnancy. The clinicians’ clinical judgment is key to starting down the road of investigation. 

To D-Dimer Or Not To D-Dimer?

In non-pregnant patients with a low pre-test probability, the D-dimer is an appropriate first step to rule-out PE. In pregnant patients, there is a physiologic increase in the D-dimer, thus the utility of this test in this population has been questioned. According to the American Thoracic Society, the recommendation is to not use a D-dimer to rule-out PE in pregnancy (4). The guidelines cite a study by Damodaram et al and found the D-dimer to have a sensitivity of 73% and a specificity of 15% with a negative likelihood ratio of 1.8 (5). However, Jeff Kline proposes that D-dimers can be of use with some alteration of “normal value.” In a podcast at ERCast.org, Dr. Kline proposed the D-dimer threshold should increase 50% from its original normal threshold for each pregnancy trimester. For example, if a normal threshold is 500ng/dL, the pregnancy-adjusted cutoffs would be: first trimester 750ng/mL, second trimester 1000ng/mL, and third trimester 1250 ng/mL (6). These adjustments are proposed for evaluation by future research, but currently have not been validated in the literature. The consensus at this time is that D-dimer is not an adequate test to rule-out PE in a pregnant patient.  

Wait, Can We PERC Out Pregnant Patients?

Unfortunately, the PERC rule has not been validated in pregnancy. Kline et al performed a retrospective analysis of 1880 patients with confirmed diagnosis of acute PE. Within this cohort, both pregnant and post-partum patients showed a statistically significant difference in their results of PERC positive and PERC negative, suggesting the PERC rule is inadequate to rule-out PE in these populations (7). Dr. Kline is currently validating a modified PERC rule for pregnant patients, however no evidence for this modified test exists at this time.

Further Investigation

The American Thoracic Society (ATS) clinical guidelines for evaluation of PE in pregnancy are the most up to date and encompassing of the current literature.  The ATS solicited recommendations of panelists from numerous experts in the field, including the presidents of ACOG and the Society of Nuclear Medicine. Panel participants included physicians who specialized in PE diagnostics, management, and treatment, especially in pregnant patients. These included cardiothoracic radiologists, nuclear medicine physicians, pulmonologists, and obstetric gynecologists.

Below is a brief summary of the guidelines with highlights of the thought-process and supporting evidence.

Recommendation 1:

D-dimer should not be used to rule-out the presence of pulmonary embolism in a pregnant patient. (See “To D-Dimer or To Not D-Dimer?” for explanation)

Recommendation 2:

If a pregnant patient has signs/symptoms of a deep venous thromboembolism (DVT), it is recommended to investigate with bilateral lower extremity venous compression ultrasound.

This recommendation stems from the common sense that if a DVT is found, anticoagulation can begin without further investigation. This scenario can benefit the patient by avoiding radiation from further imaging of a ventilation-perfusion (VQ) or a computed-tomography pulmonary angiogram (CTPA), assuming the patient is hemodynamically stable. However, with the prevalence of DVTs in pregnancy being unknown, it is recommended that ultrasound should not be performed in instances where there are no signs/symptoms of DVT.  

Recommendation 3:

In a pregnant patient with no signs/symptoms of DVT but with clinical suspicion of a pulmonary embolism, perform a chest x-ray (CXR) as the first imaging step. 

As described by Worsley and colleagues in review of the PIOPED trial, no CXR findings have been associated with the presence or absence of a pulmonary embolism (8). However, the CXR has the opportunity to point the clinician towards an alternative diagnosis for the patient’s presenting dyspnea (eg pulmonary edema, pneumonia, etc). If no alternative diagnosis is found, the CXR is an important imaging step as it can steer the clinician to the next imaging modality, VQ vs CTPA, assuming the continued work-up of PE is pursued.

Recommendation 4/5:

In a pregnant patient with a normal CXR, the next imaging recommendation is to perform a VQ scan. In a pregnant patient with an abnormal CXR, the next imaging recommendation is to perform a CTPA. 

Cahill and colleagues performed a retrospective cohort study of 304 pregnant and post-partum patients with clinical suspicion of PE. This study showed that after a normal CXR, patients were statistically more likely to obtain a diagnostic result (normal or high probability) with a VQ versus a CTPA (94% versus 70%, p<0.01). However, if the CXR was found to be abnormal, patients were statistically significant more likely to obtain a non-diagnostic VQ instead of an inconclusive CTPA (40% versus 16.4%, p<0.05), and thus the recommendation of jumping to CTPA after an abnormal CXR (3).

Recommendation 6:

In a pregnant patient with a non-diagnostic VQ, perform a CTPA.

This recommendation stems from the fact that due to high morbidity and mortality caused by PE in pregnancy, the focus should be on diagnostic certainty. Numerous studies have cited variable ranges of sensitivity and specificity of CTPA for diagnosing PE. Unfortunately, most studies exclude pregnant patients. PIOPED II, the largest study to date, also excluded pregnant patients, however the study cited a sensitivity of 83% and specificity of 96% of the CTPA for PE diagnoses (9).

Summary Of Recommended Steps From The American Thoracic Society:


Adapted from Figure 1 of the American Thoracic Society Guidelines (4).


Shared Decision Making

At the end of the day, the work-up and choice of imaging will come down to shared decision-making with the patient. The primary concern for the patient and fetus is radiation exposure from VQ scan and CTPA. While consensus is that radiation exposure is most harmful to the fetus in early pregnancy, there is debate with regards to the amount of radiation exposure from a VQ VS CTPA as the pregnancy progresses. According to a blog at ERCast.org, Dr. Kline quantifies the radiation exposure with regards to money. 0.1 Gray is analogous to $100. CTPA is worth 25-50cents. VQ is worth 50-75 cents. And the amount of radiation necessary to increase a person’s risk of cancer to 1/1000 is $10. Thus, with CTPA and VQ causing less than $1 of radiation, the mortality caused by a pulmonary embolism greatly outweighs the risk of malignancy and it is highly recommended to err on the side of further investigation (6).

Other Things To Contemplate...

tPA In A Pregnant Patient?

Common sense tells us this is a poor idea. It is widely know that pregnancy is a relative contraindication to using tissue plasminogen activator (tPA). The risk of maternal bleeding and fetal loss is the most widely feared complication. However, case reports of PE in pregnant patients successfully treated with tPA are scattered throughout the literature. 

Fasullo et al described a case of a patient at 24 weeks gestation who presented with a massive PE complicated by shock. The decision was made to administer tPA based on the patient’s hemodynamic instability. This patient stabilized within thirty minutes of administration and close obstetrical follow-up confirmed with no negative effects to the fetus. The patient was discharged without any bleeding sequelae and she delivered a healthy baby at term without fetal consequence (11). Trukhacheva et al in Obstetrics and Gynecology and Ahearn et al in Archives of Internal Medicine echoed similar scenarios (12, 13). In all cases, tPA was used as a last resort. In summary, pregnancy remains a relative contraindications to push tPA. However, the literature does demonstrate successful cases of tPA administration in pregnant patients with PE. While there insufficient data to create clinical guidelines, physicians should keep tPA in mind if all available treatment options have been exhausted.  

So, if we can’t D-Dimer or PERC out pregnancy patients… Do we have to CTPE/VQ Scan EVERY pregnant patient with dyspnea for a PE?

That’s an excellent question. The answer is, no but more so – it depends.  While PE in pregnant patients is a diagnosis that physicians must always bear in mind, a recent systemic review and meta-analysis by Kline et all challenges this dogma. He found that symptomatic ED pregnant patients actually had a low outcome rate of venous thromboembolism (VTE) and a lower relative risk of VTE than non-pregnant ED patients as clinicians employ a lower threshold for testing for pregnant patients with suspected PE (10). Thus, if there is another diagnosis that is more likely, prioritize that work-up over the PE work-up while keeping PE on the differential diagnosis. Clinical suspicion of the physician will ultimately guide appropriate diagnostic testing. Because PE in pregnancy has such high morbidity and mortality, clinicians must continue to maintain a high degree of clinical suspicion for this disease while balancing risks of diagnostic testing. 

Expert Review

Dr. Sanders did an excellent job describing this conundrum. She provides a good overview of challenges physicians face and some important literature and podcast summary data. There are a few topics were I might be able to add somewhat to her otherwise thorough summary.

    + Chicken and the egg problem:

You have to first suspect PE before deciding to test for PE….The Cahill reference suggests that there is no clinical feature associated with PE in pregnancy (3). I am not sure I would agree with this – their study was really designed to measure which approach yielded more non-diagnostic imaging (CT vs VQ? The answer….CT 17% vs. VQ 13%. This difference however was not statistically significant perhaps due to several factors including likely lack of power, which was not mentioned in their article. Dr. Sanders is right, they did try to analyze which clinical factors were possibly associated with increased likelihood of PE diagnosis; none were significant. But again this is likely due to lack of power as well as the retrospective (and perhaps less complete) manner in which chart data were abstracted. That is an easier explanation to swallow than that there are simply no clinical signs or symptoms in pregnant patients useful to start the ball rolling with respect to clinician suspicion. Dr. Sanders is right – clinicians' clinical judgment is key – but there has to be something to base this judgment on.

    + For me, it is the same data that we use to search out PE in non-pregnant patients:

Specifically, I am more likely to test patients with dyspnea either at rest or with exertion. Granted that can be difficult in later pregnancy where there are a variety of factors that can cause fatigue. I am more likely to test patients with pleuritic chest pain than vague, brief, or resolved chest pain in isolation. And I am more likely to test when symptoms are not likely explained by other factors. Until there are larger datasets of PE testing in pregnancy, my approach is largely similar to my approach in non-pregnant patients.

    + We have a deficit of pregnancy data with respect to PE:

The Kline 2013 letter to the editor mentioned by Dr. Sanders was a re-analysis of data from the EMPEROR registry (Multi-center EM PE in the Real World registry) where of 1880 patients with PE, only 14 were pregnant and 16 post partum (7). Numbers of patients in other studies were low (14,15).

    + I appreciated the ATS guidelines, but did not find them very novel or helpful:

Sure, it makes sense to start with duplex lower extremity US if you also suspect DVT and I would agree that doing this in situations where there are no signs and symptoms of DVT does not make sense. But ultimately, that whole figure boils down to either testing or not...and choosing either CT or VQ...I believe both to have roughly similar maternal radiation risk profiles as a single exposure which is significantly below a level where cancer risk should obviate the use of imaging.

    + The fetal risk exposure was not highlighted much in this blog post:

There is not much guidance from literature here, but common sense suggests avoidance of radiation during organogenesis is most critical (approximately up to the 9th week gestation). Luckily, this is when pregnant patients may be least likely to show up with fatigue, dyspnea, or non-specific chest symptoms, which are more likely in the third trimester.

    + One final thought on decision-making in pregnancy regarding the importance of communication:

Find out exactly why they are there. An OB saw them and was concerned? The patient called the OB triage nurse and was told “Come get tested for PE.” The patient read about something on the internet?... All require slightly different approaches to both your decision making as well as the need for testing and your communication approach with the patient. Also, I try in almost all situations to involve the OB physician who knows the patient the best and at times if available, the father in the discussion. Discussion with radiology MDs and techs is often needed to ensure the consent process from their end is well informed and patient-centered.

D. Mark Courtney MD MSCI
Director of Research; Associate Professor; Department of Emergency Medicine Northwestern University, Feinberg School of Medicine [Pubmed]


  1. Centers for Disease Control and Prevention. Pregnancy-related mortality surveillance –United States, 1991-1999. MMWR Morbidity Mortality Weekly Report 2003; 52:1-8. 
  2. Gherman RB, Goodwin TM, Leung B, Byrne J, Hethumumi R, Montoro M. Incidence, clinical characteristics, and timing of objectively diagnosed venous thromboembolism during pregnancy. Obstetrics Gynecology 1999; 94: 730-734. 
  3. Cahill AG, Stout MJ, Macones GA, Ghalla S. Diagnosing pulmonary embolism in pregnancy using computed-tomographic angiography or ventilation-perfusion. Obstetrics Gynecology 2009; 114: 124-129. 
  4. Leung AN, Bull TM, Jaeschke R, Lockwood CJ, Boiselle PM. Evaluation of Suspected Pulmonary Embolism in Pregnancy. American Journal of Respiratory Critical Care Medicine 2011; 184: 1200-1208. 
  5. Damodaram M, Kaladini M, Luckit J, Yoong W. D-dimers as a screening test for venous thromboembolism in pregnancy: is it of any use? Journal of Obstetrics Gynecology 2009; 29: 101-103. 
  6. Orman, Rob. "Pulmonary Embolism in Pregnancy with Jeff Kline." ERCAST.ORG. 24 Apr. 2013. Web. <http://blog.ercast.org/pulmonary-embolism-in-pregnancy/>.
  7. Kline JA, et al. Clinical Features of Patients With Pulmonary Embolism and a Negative PERC Rule Result. Ann Emerg Med. 2013 January 60(1): 122-124.
  8. Worsley DF, Alavi A, Aronchick JM, Chen JT, Greenspan RH, Ravin CE. Chest radiographic findings in patients with acute pulmonary embolism: observations from the PIOPED Study. Radiology 1993; 189: 133-136. 
  9. Stein PD, Fowler SE, Goodman LR, Gottschalk A, Hales CA, Hull RD, Leeper KV Jr, Popovich J Jr, Quinn DA, Sos TA, Sostman HD, Tapson VF, Wakefield TW, Weg JG, Woodard PK. Multidetector computed tomography for acute pulmonary embolism. New England Journal of Medicine. 2006; 354 (22):2317.
  10. Jeffrey A. Kline et al. Systematic Review and Meta-analysis of Pregnant Patients Investigated for Suspected Pulmonary Embolism in the Emergency Department. Academic Emergency Medicine. Volume 21, Issue 9, pages 949–959, September 2014.
  11. Fasullo, S. Maringhini, G. Terrazzino, G. Ganci, F. Paterna, S. Di Pasquale, P. “Thrombolysis for massive pulmonary embolism in pregnancy: a case report.” Int J of Emerg Med. 2011: 4:69. 
  12. Trukhacheva, E. Scharff, M. Gardner, M. Lakkis, N. “Massive pulmonary embolism in pregnancy treated with tissue plasminogen activator.” Obstet Gynecol. 2005 Nov; 106 (5 Pt 2): 1156-8. 
  13. Ahearn GS, Hadjiliadis D, Govert JA, Tapson VF. Massive Pulmonary Embolism During Pregnancy Successfully Treated With Recombinant Tissue Plasminogen Activator: A Case Report and Review of Treatment Options. Arch Intern Med. 2002;162(11):1221-1227. 
  14. Courtney DM, Kline JA, Kabrhel C, et al. Clinical features from the history and physical examination that predict the presence or absence of pulmonary embolism in symptomatic emergency department patients: results of a prospective, multicenter study. Annals of Emergency Medicine. 2010;55(4):307–315.e1. doi:10.1016/j.annemergmed.2009.11.010.
  15. Kline JA, Courtney DM, Kabrhel C, et al. Prospective multicenter evaluation of the pulmonary embolism rule-out criteria. J Thromb Haemost. 2008;6(5):772-780. doi:10.1111/j.1538-7836.2008.02944.x.
Posted on January 19, 2016 and filed under Pulmonary.