Pulmonary embolism in pregnancy with Jeff Kline

What is the best way to evaluate a pregnant patient in whom you suspect pulmonary embolism? There is no definitive guideline, but there is no shortage of opinions. Jeff Kline and I work through the evidence and lack thereof.


What is the fetal radiation exposure from CT pulmonary angiogram and ventilation perfusion scan (V/Q)?

What do you tell pregnant patients about how much fetal radiation there is from a radiographic study? Do you use rads, grays, seiverts, micrograys? Here’s any easy way to think about it… The threshold we want to avoid is 0.1 gray. 0.1 gray at any time during gestation is regarded as the practical threshold beyond which induction of congenital abnormalities is possible. Do you know how much 0.1 gray is in relation to rest of the universe? Don’t worry, nobody else does either. To help with perspective, think of 0.1 gray as $100 or 100 points, we’ll use dollars here.

Estimated fetal radiation exposure from diagnostic imaging studies

0.1 gray = $100

A chest x-Ray is one tenth of a penny.

Ct pulmonary angiogram is 25 to 50 cents.

A V/Q scan is 50 to 75 cents, less with a partial dose V/Q, which is often used in pregnancy. So for CT and V/Q , we’ll say 50 cents each.

Background radiation during 9 months gestation: $5.

Amount of radiation to increase the risk of cancer before age 20 by one one-hundredth of a percent, or one in ten thousand, $10.

On the scale of $100, CT and VQ give less than $1 of fetal radiation exposure. Considering the risk of a bad outcome from PE, especially in a pregnant patient, where it is one of the leading causes of maternal mortality, err on the side of the workup.

  • In pregnancy,  there is a higher chance of a non-diagnostic CTPA (35%) vs V/Q (4%).
  • With an abnormal chest x-ray, the utility of V/Q is diminished and CTPA is a better study.
  • The amount of breast radiation is markedly higher for a CTPA than for a V/Q scan. In a 25 year old woman, it is estimated that a CT chest increases the lifetime risk of breast cancer by 1.5%.

So what study to do? My approach is to perform a V/Q scan in a pregnant patient with a normal chest x-ray. The caveat to this is early in the first trimester, where the decision may be more emotion than data based (on my part). At this stage of gestation, the  fetus is about the size of a cashew nut and, with a V/Q. retained urine in the bladder seems like a lot of focused radiation to the entire fetus. In this patient group, I start with CTPA.

Jeff Kline’s approach to using the D-dimer in pregnancy

We don’t always get a d-dimer when pulmonary embolism is on the differential diagnosis (at least I hope not), but it can be helpful in some cases. In normal pregnancy, 60% of patients will have a d-dimer above standard threshhold, which limits the test’s clinical utility. Jeff suggests using higher d-dimer cutoffs as pregnancy progresses – 50% higher for each trimester.  If your regular cutoff is 500 ng/mL, pregnancy corrected cutoffs:

  • First trimester 750 ng/mL
  • Second trimester 1000 ng/mL
  • Third trimester 1250 ng/mL

As discussed in the podcast, Jeff also uses this approach in the first trimester:  Negative PERC rule (with the caveat of an increased heart rate of 105) plus a negative d-dimer makes PE unlikely. This uses  PERC like a low risk Well’s score, but is not how PERC should be used in non-pregnant patients. To date, there is no published data set that validates this strategy of using PERC + negative d-dimer in pregnancy.

Tom Deloughery’s Protocol for Rivaroxaban for Venous Thromboembolic Disease


  • Acute lower extremity venous thrombosis
    • Consider catheter directed thrombolysis for large proximal vein thrombosis (editor comment: may decrease the chance of post-phlebitic syndrome)
  • Acute pulmonary embolism
    • Consider LMWH if patient hypotensive (editor comment: Tom is not a big fan of thrombolytics in PE. If a patient is hypotensive, I would strongly consider thrombolytics in properly selected pateints. We will have extensive analysis and coverage of thrombolytics for PE later in the year)


  • Creatinine Clearance  < 30 ml/min
  • Cirrhosis
  • Severe untreated hypertension (SBP > 180 or DBP > 110)
  • Patient on protease inhibitors or “azole” drugs
  • Cancer (LMWH drugs of choice)
  • Pregnancy


  • Rivaroxaban 15mg po bid x 3 weeks then 20mg daily


Paraspinous block for headache

Selected reading

Pulmonary Embolism in Pregnancy Lancet 2010. Higher rate of non-diagnostic CTPA than V/Q in pregnant patients

CT alone versus CT plus lower extremity ultrasound

D-dimer levels in normal pregnancy. Article by Jeff Kline et al

Laboratory studies in pregnancy. Obst Gyn 2009



  1. Brian F

    I haven’t listened to the podcast yet, so sorry if this is a misinterpretation, but it looks like above you are saying that you need both a negative ddimer and negative PERC to r/o a PE in pregnancy. That seems like a high standard to meet. My current practice is to basically use either to r/o PE in Pregnancy, and to really only go that far, in the 1st trimester, if something else piques my interest in the patients history (family hx of unprovoked vte, recent immobilization or a good story)

    1. Rob Orman

      Brian- this point is explained in the podcast, but I concur with your thought process. I have re-written that segment of the podcast to clarify.

    2. James Garvey, Acute Medicine Registrar, Norwich, UK

      I find pregnant patients with possible pulmonary embolism very frustrating.
      I love the idea of Kline’s approach of adjusted d-dimer thresholds. However it seems there is very little support for this in the literature. The article by Kline et al that you link to above contains this sentence: “We reemphasize that the cutoff values of 0.75, 1.00, and 1.50 mg/L are not valid in any way for clinical practice, but we submit that these thresholds could be evaluated in a prospective research study in which pregnant patients with possible venous thromboembolism are evaluated by a validated diagnostic algorithm”.
      Up to date and this review article in NEJM (http://www.nejm.org/doi/pdf/10.1056/NEJMra0707993) recommend not using d-dimer at all in assessing for PE. Up to date quotes this paper ‘A negative D-dimer does not exclude venous thromboembolism (VTE) in pregnancy’.
      The PERC rule has not been validated in Pregnancy (http://www.aliem.com/when-perc-rule-fails/).

      So Jeff Kline’s approach, while attractive, is probably not a high enough standard. It is not supported by published literature -- I’d hate to be referring to a podcast in Coroners Court. As to Brian F’s approach: not clear it is safe at all. If you ever had serious complications or a death from a missed PE in a pregnant patient (healthy, young, etc.) you’d want to be sure to have some serious literature or guidelines behind you.

      If Jeff is reading this -- please do some more work, publish more on pregnancy and PE. At my hospital pleuritic chest pain and pregnancy is usually a direct ticket to radiation, and I’m sure we are doing more harm than good.

      1. Rob Orman

        Agreed, James. Jeff and I discussed this at Essentials of Emergency Medicine last year (can be found in the digital download) as well as on an upcoming EMRAP (probably early 2015). There is literature supporting increased cutoffs for d-dimer in pregnancy, but no consistent data showing that these are the hard cutoffs.

  2. Pingback: PE Prognostication with Dr Anand Senthi - Broome Docs

  3. Pingback: PE Part 1 | EM Basic

  4. Pingback: Investigation of PE in pregnancy

Awesome article, I know - please share your erudite thoughts...