Learn EM

Pulmonary Embolism Algorithm

The American College of Emergency Physicians recently revised its clinical policy on the evaluation of adults presenting to the emergency department with suspected acute venous thromboembolic disease. It was published in the May issue of Annals of Emergency Medicine and is available to read here. The revision answers 5 key questions regarding the evaluation and treatment of patients with thromboembolic disease and is an excellent resource. A summary for use at the bedside can be found here: (Sam’s Notes)

A review of this clinical policy and current evidence allows for development of a decision algorithm for rapid assessment of a patient with suspected pulmonary embolism. Much has been published already on the subject. Notably, Cam Berg has spoken about his work at North Memorial Health, developing their accelerated diagnostic pathway for PE. In addition, excellent resources exist at MDCalc (Wells, PERC) for calculation of risk and review of evidence.

Here is one such algorithm that is supported by current evidence and clinical policy recommendations.

Text version of above algorithm:

  1. Suspect PE (if not, stop)
  2. Use Gestalt or Wells to stratify into low or high risk
    • Low risk –  PERC
      • If PERC negative – stop
      • If PERC positive, obtain D-dimer
        • D-dimer negative, stop
        • D-Dimer positive,  obtain CTPA
    • High Risk – Obtain imaging with CTPA


In 2001, the first Wells study was published1, providing us with a tool to stratify patients with suspected pulmonary embolism into three risk categories: low, intermediate, and high. The tool was validated in the emergency department and then re-studied with two risk categories2,3 instead, in order to simplify the model. It was found to be equally effective utilizing a low/high risk categorization with low risk patients receiving d-dimer testing. Low risk patients with a negative d-dimer were discharged home without further testing. The groups were found to have 3% vs 28% prevalence of VTE.2

In 2004, the PERC (Pulmonary Embolism Rule-out Criteria)4 rule was published. The intent was to determine a set of criteria that could be use to identify low risk patients who do not need any further testing for PE (ie, no d-dimer). The PERC rule required two steps. First identify patients who are low risk for PE (pre-test probability <15%) by either physician gestalt or applying the wells criteria. Second, apply the PERC rule to this population and if negative, avoid any further testing for PE. This rule was also validated in the ED.5,6

Since then, studies have been published describing age based cut-offs for d-dimer results, which have been adopted by the 2018 ACEP guidelines for screening patients with possible venous thromboembolism (VTE).7,8,9 The recommended age-based cut offs are:

  • D-dimmer cut off equal to age after 50 (age x 10 ug/L FEU or age x 5 ug/L DDU)
  • D-dimer cutoff that increases by decade
  • Set higher d-dimer cut off for patients between 50-70

The algorithm published here utilizes all of these published evidence and guidelines to achieve the simplest path to screening a patient with possible PE. Keeping in mind that the PERC rule requires a patient to be “low risk” prior to applying the rule, the wells criteria are recommended. However, there is room for physician gestalt. If the PERC rule is negative, these patients have a risk of VTE <1.5%.5,6 Patient that are deemed “high risk” have imaging performed without waiting for d-dimer results.

This algorithm is one option for evaluating patients with possible PE. Other protocols have been developed utilizing these same criteria and the original wells 3 tier strategy. Most notably, Dr. Cam Berg has discussed his accelerated diagnostic protocol for PE. Further information on his ADP can be found here: EM:RAP

One important note from the Dr. Kline and the original PERC rule team. The validation study of the PERC rule5 had a prevalence of PE/DVT of 6%. It suggests that a setting with higher prevalence may not be able to effectively screen low risk patients by PERC and that independent validation should occur at the specific site before complete adoption.

Algorithm FAQ:

  1. Why the double scoring systems? Do we really need a Wells AND a PERC? Yes. The PERC rule has gained popularity as the only rule needed to exclude a PE in a patient clinically, but there is an important caveat. The patient must be “low risk” before applying the PERC. How do we make that determination? Many of us use clinical gestalt, which has been proven to be adequate in the 2008 PERC validation study5. However, many of the criteria we use for our gestalt come from the validated Wells criteria. In either case, we acknowledge that there is a pre-PERC determination of risk. This is why this algorithm lists two calculations. A moderate or high risk patient is not appropriate for the PERC criteria and was not in the studied population.
  2. There is significant overlap between the Wells and PERC criteria. Doesn’t this represent double dipping if we make the same calculation twice? No, not exactly. The overlap between the two calculations is clear, but they are not identical and they were not studied to make the same determination. Wells was studied to apply a d-dimer and PERC was studied to screen low risk patients and obviate any testing at all.
  3. Do I need any data from my institution before applying this rule? Yes. Knowing the prevalence of PE/DVT in your ED population is helpful. As stated in the 2008 PERC validation study, their prevalence was 6.9% for PE & DVT combined. The discussion notes that “the underlying prevalence of VTE may be too high in other ED populations for this strategy to be able to safely rule out PE”.
  4. If a patient is PERC positive, why don’t I just go directly to CTPA? The PERC rule is unidirectional. If negative, it is excellent at determining risk <1.8% (as low as 1.3%4,5,6) and obviating further testing. However, if positive, it has no predictive value. In that scenario we look back at the previous question, pre-test probability by Wells, and see that the patient was felt to be low risk. That qualifies the patient to receive a d-dimer which if negative, allows us to rule out PE without imaging.
This algorithm is published with the intent of providing a structure for the evaluation of patients with suspected PE. As stated in the validation studies, reviews, and published recommendations, protocols are not meant to replace physician judgement which has been validated as an important diagnostic tool.

Sam’s note for use at bedside: PE algorithm


  1. Wells PS, Anderson DR, Rodger M, et al. Excluding pulmonary embolism at the bedside without diagnostic imaging: management of patients with suspected pulmonary embolism presenting to the emergency department by using a simple clinical model and d-dimer. Ann Intern Med. 2001;135(2):98-107. PubMed
  2. Wolf SJ, Mccubbin TR, Feldhaus KM, Faragher JP, Adcock DM. Prospective validation of Wells Criteria in the evaluation of patients with suspected pulmonary embolism. Ann Emerg Med. 2004;44(5):503-10. PubMed
  3. Van belle A, Büller HR, Huisman MV, et al. Effectiveness of managing suspected pulmonary embolism using an algorithm combining clinical probability, D-dimer testing, and computed tomography. JAMA. 2006;295(2):172-9. PubMed
  4. Kline JA, Mitchell AM, Kabrhel C, Richman PB, Courtney DM. Clinical criteria to prevent unnecessary diagnostic testing in emergency department patients with suspected pulmonary embolism. J Thromb Haemost. 2004;2(8):1247-55. PubMed
  5. Kline JA, Courtney DM, Kabrhel C, et al. Prospective multicenter evaluation of the pulmonary embolism rule-out criteria. J Thromb Haemost. 2008;6(5):772-80. PubMed
  6. Freund Y, Cachanado M, Aubry A, et al. Effect of the Pulmonary Embolism Rule-Out Criteria on Subsequent Thromboembolic Events Among Low-Risk Emergency Department Patients: The PROPER Randomized Clinical Trial. JAMA. 2018;319(6):559-566. PubMed
  7. Raja AS, Greenberg JO, Qaseem A, et al. Evaluation of Patients With Suspected Acute Pulmonary Embolism: Best Practice Advice From the Clinical Guidelines Committee of the American College of Physicians. Ann Intern Med. 2015;163(9):701-11.  PubMed
  8. Wolf SJ, Hahn SA, Nentwich LM, et al. Clinical Policy: Critical Issues in the Evaluation and Management of Adult Patients Presenting to the Emergency Department With Suspected Acute Venous Thromboembolic Disease. Ann Emerg Med. 2018;71(5):e59-e109. Link
  9. Righini M, Van es J, Den exter PL, et al. Age-adjusted D-dimer cutoff levels to rule out pulmonary embolism: the ADJUST-PE study. JAMA. 2014;311(11):1117-24. PubMed

2 thoughts on “Pulmonary Embolism Algorithm

  1. Hi Sam! I really like your algorithm because it is so simple and straight forward. However, I’m not sure you should be applying the PERC rule to patients with a Well’s score of 2-4. These patients are actually intermediate risk and thus PERC should not be applied to them. Has there been a study showing that PERC can be safely applied to patients who are deemed “PE unlikely” using the simplified Well’s criteria? I am not aware of one.

    1. Thanks JP. You’ll see, in the description above, a reference to the validation studies of Wells. Though the original validation was 3 tiered, there has been subsequent validation for a two tier categorization instead (citation above). The low/high risk stratification gives a low risk that is well below the required 15% or less required by PERC. Recall that the original PERC validation asked providers if a patient had a 15% or less risk of PE based on gestalt or Wells. Since a two tiered approach still falls below that cut off, one can safely use low/high and include patients up to 4 by Wells as low risk. Thus this algorithm. Hope that helps!

Leave A Comment