Clinical article · Acute care, Specialties, Medicines and results

Suspected pulmonary embolism: pre-test probability, D-dimer and when to image

Written for
Hospital physicians, emergency doctors and residents
Reading time
8 minutes
Published
Updated

Written by the Sana Health team for clinicians. It summarises published guidance, is not a guideline and has not been peer reviewed. Check current local protocols before acting.

Key points

  • Every work-up for pulmonary embolism starts with an explicit pre-test probability; the D-dimer is interpreted only in that light.
  • In patients with a low clinical probability, a negative PERC rule means no D-dimer is needed.
  • Age-adjusted and probability-adjusted D-dimer thresholds, such as YEARS, safely reduce imaging but must be applied as validated.
  • Patients with a high clinical probability go straight to imaging, and anticoagulation can be started while waiting if bleeding risk allows.
  • In pregnancy, pulmonary embolism is a leading cause of maternal death and needs a dedicated, adapted pathway rather than avoidance of testing.

Why a structured approach

The symptoms of pulmonary embolism overlap with those of many other conditions, so clinicians test often. Overuse of CT pulmonary angiography and of D-dimer adds radiation, contrast exposure, incidental findings and cost without improving care [3]. A structured pathway avoids both missed emboli and unnecessary scans, and its best form depends on the setting and the local prevalence of disease [11].

The 2019 ESC guideline [1] and the 2026 AHA/ACC guideline [2] follow the same sequence: estimate clinical probability, use a rule-out step or a D-dimer only where the probability allows, and image the remainder. The unstable patient with suspected pulmonary embolism is the exception: bedside echocardiography and resuscitation come first, and the work-up below does not apply.

Step 1: clinical probability

Use a validated score or structured gestalt. The Wells score is the most widely used; the revised Geneva score is an alternative that does not depend on the clinician's judgement of the likeliest diagnosis. The simplified Wells rule gives one point per item, and combined with an age-adjusted D-dimer it can safely rule out pulmonary embolism [6].

Wells itemPoints
Clinical signs and symptoms of deep vein thrombosis3
Pulmonary embolism is the most likely diagnosis3
Heart rate above 1001.5
Immobilisation for 3 days or more, or surgery in the previous 4 weeks1.5
Previous deep vein thrombosis or pulmonary embolism1.5
Haemoptysis1
Active cancer1

In the two-level version, a total above 4 means pulmonary embolism is likely and 4 or less means it is unlikely. Patients in whom it is likely go directly to imaging without a D-dimer.

Step 2: PERC in the low-probability patient

The pulmonary embolism rule-out criteria (PERC) are eight items: age 50 or over, heart rate 100 or more, oxygen saturation below 95%, unilateral leg swelling, haemoptysis, recent surgery or trauma, previous venous thromboembolism, and oestrogen use. In a patient whose clinical probability is already low, the absence of all eight means no further testing is needed. In the PROPER trial in French emergency departments, a PERC-based strategy in low-probability patients was non-inferior to conventional testing for subsequent thromboembolic events [4].

PERC is only valid when clinical probability is low before it is applied. Using it to talk down a patient with a moderate probability removes its safety.

Step 3: the D-dimer threshold

A negative D-dimer excludes pulmonary embolism in patients who are not high probability. The conventional cut-off of 500 micrograms/L loses specificity with age, and an age-adjusted cut-off of age multiplied by 10 micrograms/L in patients over 50 improves specificity while keeping sensitivity [5, 6]. Probability-adjusted strategies raise the threshold further when clinical probability is low.

StrategyWho it applies toD-dimer that excludes PE
ConventionalPE unlikely by Wells or GenevaBelow 500 micrograms/L
Age-adjustedPE unlikely and age over 50Below age x 10 micrograms/L
YEARSNo YEARS item (signs of DVT, haemoptysis, PE most likely)Below 1000 ng/mL
YEARSOne or more YEARS itemsBelow 500 ng/mL

Guidelines do not agree on a single threshold [7]. Probability-adjusted thresholds make the test more efficient but, applied outside validated strategies, may raise the failure rate, so they should be used as the studies used them [8]. The YEARS algorithm has also been tested against CT alone in a randomised trial in patients with cancer, a group for whom guidelines have generally advised direct imaging [9]. D-dimer units and assays differ: know whether the local laboratory reports fibrinogen-equivalent units.

Step 4: imaging

  • CT pulmonary angiography is the first-line test for most patients.
  • Ventilation-perfusion scanning is an alternative when contrast is contraindicated, in younger patients and in pregnancy, ideally with a normal chest radiograph.
  • Leg compression ultrasound in a patient with symptoms of deep vein thrombosis can confirm venous thromboembolism and avoid chest imaging.
  • In the unstable patient who cannot go to CT, right ventricular dysfunction on bedside echocardiography supports treating as high-risk pulmonary embolism [1].
  • In high-probability patients, anticoagulation may be started while imaging is awaited if bleeding risk is acceptable.

Once embolism is confirmed, risk stratification with PESI or simplified PESI, right ventricular assessment and troponin guides outpatient care, admission or reperfusion [1, 12]; the 2026 AHA/ACC guideline introduces new clinical categories of severity for this purpose [2].

Pregnancy

Pulmonary embolism is one of the leading causes of pregnancy-related death in high-income countries, and deaths have been linked to delayed recognition and investigation [10]. The revised Geneva score and the pregnancy-adapted YEARS algorithm, combined with D-dimer and leg ultrasound when there are leg symptoms, allow some pregnant patients to avoid chest imaging [10]. When imaging is needed, the fetal radiation dose from either CT or perfusion scanning is low, and the choice follows local obstetric and radiology protocols. Investigation should not be withheld because the patient is pregnant.

Traps

  • Requesting a D-dimer before deciding on clinical probability, then imaging every positive result.
  • Using a D-dimer at all in a high-probability patient.
  • Applying PERC to a patient who is not low probability.
  • Forgetting the age adjustment in older patients, which leads to avoidable scans.
  • Missing pulmonary embolism behind an apparent exacerbation of COPD or heart failure.
  • Sending an unstable patient to the CT scanner.

How Sana Health helps

From the history, observations and any D-dimer result, the Respiratory Medicine engine estimates clinical probability with Wells or Geneva, checks whether PERC applies, and interprets the D-dimer against the conventional, age-adjusted or YEARS threshold that fits, showing the calculation. It recommends the imaging that suits the patient, including in pregnancy and renal impairment, and once embolism is confirmed applies PESI or sPESI with right ventricular and troponin findings. Its references come from PubMed and ClinicalTrials.gov.

Questions

How is the age-adjusted D-dimer calculated?

For patients over 50, the cut-off is the age in years multiplied by 10 micrograms/L (fibrinogen-equivalent units). A 72-year-old's threshold is 720 micrograms/L. It applies only to patients in whom pulmonary embolism is unlikely by a validated score.

Can PERC be used in every patient with chest pain or breathlessness?

No. PERC should only be applied when the clinician already judges the probability of pulmonary embolism to be low. If any of the eight criteria is present, or the probability is not low, the next step is a D-dimer or imaging.

Should a pregnant patient with suspected pulmonary embolism have a CT?

If an adapted pathway with D-dimer and, where relevant, leg ultrasound does not exclude it, yes or a perfusion scan, according to local protocol. The radiation dose to the fetus is low, and missing pulmonary embolism is the greater risk.

References

  1. 2019 ESC Guidelines for the diagnosis and management of acute pulmonary embolism developed in collaboration with the European Respiratory Society (ERS) (opens in a new tab)Eur Heart J, 2020
  2. 2026 AHA/ACC/ACCP/ACEP/CHEST/SCAI/SHM/SIR/SVM/SVN Guideline for the Evaluation and Management of Acute Pulmonary Embolism in Adults: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines (opens in a new tab)Circulation, 2026
  3. Evaluation of Patients With Suspected Acute Pulmonary Embolism: Best Practice Advice From the Clinical Guidelines Committee of the American College of Physicians (opens in a new tab)Ann Intern Med, 2015
  4. Effect of the Pulmonary Embolism Rule-Out Criteria on Subsequent Thromboembolic Events Among Low-Risk Emergency Department Patients: The PROPER Randomized Clinical Trial (opens in a new tab)JAMA, 2018
  5. Diagnostic accuracy of conventional or age adjusted D-dimer cut-off values in older patients with suspected venous thromboembolism: systematic review and meta-analysis (opens in a new tab)BMJ, 2013
  6. The original and simplified Wells rules and age-adjusted D-dimer testing to rule out pulmonary embolism: an individual patient data meta-analysis (opens in a new tab)J Thromb Haemost, 2017
  7. D-dimer Levels for the exclusion of pulmonary embolism: making sense of international guideline recommendations (opens in a new tab)J Thromb Haemost, 2024
  8. Clinical probability-adjusted (CPA) D-dimer thresholds for the exclusion of venous thromboembolism (VTE) in the emergency department (opens in a new tab)Emerg Med J, 2025
  9. YEARS Algorithm for Diagnosis of Suspected Pulmonary Embolism in Patients With Cancer: A Randomized Clinical Trial (opens in a new tab)JAMA, 2026
  10. Diagnosing Pulmonary Embolism During Pregnancy (opens in a new tab)Chest, 2025
  11. Ruling out pulmonary embolism across different healthcare settings: A systematic review and individual patient data meta-analysis (opens in a new tab)PLoS Med, 2022
  12. Pulmonary embolism: update on management and controversies (opens in a new tab)BMJ, 2020