Date: November 13th, 2019

Reference: van der Pol et al. Pregnancy-Adapted YEARS Algorithm forDiagnosis of Suspected Pulmonary Embolism. NEJM 2019

Guest Skeptic: Dr. Theresa Robertson-Chenier is currently an Emergency Physician practicing at the Peterborough Regional Health Centre. She is also an adjunct faculty member with Queen’s University, Department of Family Medicine.

Case: A 32-year-old female, G1P0, who is 22 weeks pregnant, presents to your local emergency department with the chief complaint of shortness of breath. She states that for the last one week she has had progressive shortness of breath on exertion. She denies any chest pain, fever, cough or leg swelling. She has no history of venous thromboembolic (VTE) disease like deep vein thrombosis (DVT) or pulmonary embolism (PE). But recently she drove seven hours from London, Ontario to Montreal, Quebec. She is worried about the possibility of a PE. She is otherwise healthy, takes only prenatal vitamins and has no allergies. She is terrified about any radiation exposure in pregnancy and has read on google that there is a blood test you can order to rule out PE.

Background: We have covered VTE a number of times on the SGEM. This has even included a few of episodes with the PE guru and PERC rule creator Dr. Jeff Kline. However, we have never looked at the YEARS criteria study published by Van der Hulle T et al (The Lancet 2017).

  • SGEM#51: Home (Discharging Patients with Acute Pulmonary Emboli Home from the Emergency Department)
  • SGEM#126: Take me to the Rivaroxaban – Outpatient treatment of VTE
  • SGEM#163: Shuffle off to Buffalo to Talk Thrombolysis for Acute Pulmonary Embolism
  • SGEM#219:Shout, Shout, PERC Rule Them Out

The YEARS algorithm starts with the clinician suspecting an acute PE. Then they order a D-dimer and apply the YEARS clinical decision instrument. It has three items with each getting one point:

  1. Clinical signs of DVT
  2. Hemoptysis
  3. PE most likely diagnosis

If there are zero YEARS items and the d-dimer is <1,000ng/ml then a PE is excluded. If there are zero YEARS items but the d-dimer is equal to or greater than 1,000ng/ml then a CT pulmonary angiography (CTPA) scan is needed to rule out a PE.

If there are one or more YEARS items and the d-dimer is <500ng/ml then a PE is excluded. If there are one or more YEARS items but the d-dimer is equal to or greater than 500ng/ml then a CTPA scan is needed to rule out a PE.

While this publication was interesting, it was a prospective observational study from the Netherlands. There was a study by Kabrhel et al (AEM 2018) that was done in 17 hospitals in the USA. They compared usual care for possible PE vs. YEARS criteria. They enrolled 1,789 patients and 84 (4%) had a PE. Using standard d-dimer criteria, 53% would not have been imaged (2 misses). YEARS avoided imaging in 67%, but had 6 misses. Standard care had a sensitivity 97.6% vs, 92.9% for YEARS. It would be better if there was a randomized control trial comparing usual care to YEARS. In addition, the case you presented was of a pregnant woman. In the original YEARS study from 2017 it said pregnancy was an exclusion.

Clinically, it can be difficult to diagnosis PE in pregnancy because of the overlap of symptoms due to the physiological changes in pregnancy (tachycardia, shortness of breath and leg swelling) with the signs and symptoms of PE. The incidence of PE is reported to be 1.72 cases per 1,000 deliveries, and it accounts for approximately one death in every 100,000 deliveries.

In addition, the diagnostic tests used to diagnosis PE come with their own risks to mom and fetus. The radiation dose to the maternal breast can be potentially carcinogenic owing to the radiosensitive nature of the glandular breast during pregnancy. A CTPA study can increase the risk of breast cancer by 1.5% in a 25-year-old woman (see reference on last page).

Clinical Question: Can the YEARS algorithm, which utilizes the D-dimer test, be used in pregnant women to rule out the diagnosis of pulmonary embolism?

Reference: van der Pol et al. Pregnancy-Adapted YEARS Algorithm forDiagnosis of Suspected Pulmonary Embolism. NEJM 2019

  • Population: Pregnant women, 18 years of age and older, with clinically suspected PE (defined as new onset or worsening dyspnea, +/- hemoptysis or tachycardia) referred to the ED or the obstetrical ward.
    • Exclusions: Treatment with a full-dose therapeutic anticoagulant agent, can’t follow-up, allergy to the contrast dye, or a life expectancy of less than three.
  • Intervention: Application of the pregnancy-adapted YEARS algorithm to rule out PE in pregnant women.
  • Comparison: Not using pregnancy-adapted YEARS (Hypothetical situation in which all patients undergo CTA or VQ scan)
  • Outcome:
    • Primary Outcome: The cumulative incidence of symptomatic VTE, with confirmation by objective tests, during a 3-month follow-up period in the subgroup that anticoagulation treatment was withheld.
    • Secondary Outcome: Proportion of patients in whom CTPA was not indicated.

The pregnancy-adapted YEARS algorithm is the same as the YEARS algorithm but if the pregnant patient has signs of a DVT you get an ultrasound of the leg. If it shows a DVT you treat for VTE. If it does not show a DVT then you enter the regular YEARS algorithm.

Image from MDCalc.

Authors’ Conclusions: Pulmonary embolism was safely ruled out by the pregnancy-adapted YEARS diagnostic algorithm across all trimesters of pregnancy. CT pulmonary angiography was avoided in 32 to 65% of patients”.

Quality Checklist for Clinical Decision Tools:

  1. The study population included or focused on those in the ED. Unsure
  2. The patients were representative of those with the problem. Yes
  3. All important predictor variables and outcomes were explicitly specified. Yes
  4. This is a prospective, multicenter study including a broad spectrum of patients and clinicians (level II). Yes
  5. Clinicians interpret individual predictor variables and score the clinical decision rule reliably and accurately. No
  6. This is an impact analysis of a previously validated CDR (level I). No
  7. For Level I studies, impact on clinician behavior and patient-centric outcomes is reported. No
  8. The follow-up was sufficiently long and complete. Yes
  9. The effect was large enough and precise enough to be clinically significant. Unsure

Key Results: They recruited 498 consecutive pregnant women with clinically suspected PE into the study. The mean age was 30 years and almost half (46%) were in their third trimester. Half of the patients had no YEARS criteria and half had at least one of the three criteria (19% signs of DVT, 8% hemoptysis and 89% PEs were the most likely diagnosis).

Only one symptomatic DVT was diagnosed in follow-up that was not anticoagulated.

  • Primary Outcome:VTE at three months in the subgroup anticoagulation was withheld.
    • 477/498 (96%) VTE was ruled out at baseline
    • Only 1 DVT (0.21%)was identified during follow-up (95% CI; -0.04 to 1.2)
  • Secondary Outcomes: Proportion of patients in whom CTPA was not indicated
    • 195 patients were ruled out based on the adapted YEARS algorithm
    • 12 (6.2%) patients had a CTPA even though not indicated (protocol violation). All 12 were negative for PE.
    • CTPA could safely be avoided in 39% of the patients (95% CI 35-44)

1. Incorporation Bias: This can occur when results of the test under study are actually used to make the final diagnosis. In this study the authors acknowledge that the physician may have been aware of the d-dimer results when assessing the YEARS criteria. This can make the test (diagnostic algorithm) appear more powerful by falsely raising the sensitivity and specificity.

2. Partial Verification Bias (Referral Bias or Work-Up Bias): This can happen when only a certain set of patients who underwent the index test is verified by the reference standard. In YEARS, only those with a positive d-dimer (>1,000 or >500 depending on zero or 1+ criteria) got the definitive test. This could increase sensitivity but decreases specificity.

3. Differential Verification Bias (Double Gold Standard): This is very similar to partial verification bias and could be part of incorporation bias. Differential verification bias can occur when the test results influence the choice of the reference standard. So, a positive index test gets an immediate/gold standard test (CTPA in this case) whereas the patients with a negative index test get clinical follow-up for disease. This can raise or lower sensitivity and specificity.

4. Subjectivity: One of the weaknesses of this study is that it includes the subjective part of the Well’s criteria as part of the YEARS criteria. The part where the clinician needs to use clinical gestalt and decide if a PE is the most likely diagnosis.

5. Spectrum Bias: You made me think of one more potential bias, spectrum bias. Sensitivity depends on the spectrum of disease, while specificity depends on the spectrum of non-disease. So, you could falsely raise the specificity if the YEARS algorithm is used as a screening test. Just because a pregnant patient has some vague chest pain or shortness of breath does not get them into the algorithm. The clinician had to have a clear suspicion of PE. The best paper on these biases was Understanding the Direction of Bias in Studies of Diagnostic Test Accuracy (Kohn et al AEM 2013).

Comment on Authors’ Conclusion Compared to SGEM Conclusion: We generally agree with the authors’ conclusion.

SGEM Bottom Line: The pregnancy-adapted YEARS algorithm has the potential to safely rule out PE and decrease CTPA studies but requires external validation.

Case Resolution: Your clinical gestalt is she does not have a VTE and you do not work her up for a PE.

Clinically Application: The data on the pregnancy-adapted YEARS algorithm is encouraging but should go through external validation prior to implementation.

Dr. Theresa Robertson

What Do I Tell My Patient? You tell her that based on over 20 years of experience you do not think she has a blood clot in her lung. You explain that the blood test she googled is called a d-dimer. While it is good for ruling out blood clots it can be falsely elevated. This can lead to unnecessary CT scans with radiation that she is terrified about being exposed to. You provide reassurance and tell her to return to the ED if she gets increasing shortness of breath, develops chest pain, starts cough up blood, one leg swells up or she is worried.

Keener Kontest: Last weeks’ winner was Mario Pinoli. This is a repeat win for Mario. He knew Karl Dussik is credited with the first use of ultrasound as a diagnostic tool? Mario says I do not need to send him another cool skeptical prize as long as I pronounce his name correctly.

Listen to the podcast to hear this weeks’ trivia question. If you know the answer, send an email to with “keener” in the subject line. The first correct answer will receive a cool skeptical prize.

Other FOAMed resources:

Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.