I. How to Use

When to Use

The PERC rule can be applied when considering the diagnosis of PE in low-risk patient. A patient can be determined low-risk if their pre-test probability for PE is <15% as determined by physician gestalt. Some advocate for applying Well’s Score or Revised Geneva Score (RGS) for PE first and obtaining a score considered to be low-risk by those scoring tools.

Pearls / Pitfalls

Pearls

  • Apply only to low-risk patients (<15% by clinician gestalt). Some advocate using Wells’ or RGS to determine PE risk, but the original PERC score was intended to be used in conjunction with physician gestalt.

  • Validated only in populations where PE prevalence <7% (typical U.S. prevalence ~6%).

  • Hypoxemia (SpO2 <95% on room air) or tachycardia (HR ≥ 100) at any point during the visit counts as positive.

  • The test is unidirectional. Meeting all PERC rule out criteria allows the clinician to avoid further testing, however, failing the rule does not mean the clinician MUST order tests.

  • This tool is NOT meant to risk-stratify patients.

Pitfalls

  • Do NOT use this in patients who are at moderate or high-risk of PE OR in high-prevalence settings, as it has reduced sensitivity.

  • Caution in:

    • Pleuritic, focal chest pain (higher false-negative rate).

    • Patients with ECG evidence of RV strain from pulmonary hypertension (sinus tachycardia, S1Q3T3, RBBB, T-wave inversion in V2 and V3, ST elevation aVR, atrial fibrillation).

    • Pregnant or pediatric patients (not validated—see PERC-Peds for children or YEARS criteria).

    • Personal or family history of thrombophilia

  • Beta-blockers may mask tachycardia, falsely lowering risk.

Why to Use

  • Low-risk patients who are PERC negative avoid the risks of unnecessary testing and treatment for PE.

  • Clinicians can avoid unnecessary and potentially costly or harmful testing and treatment based on clinical criteria alone.

II. Next Steps

Advice

PERC has reduced sensitivity in high-prevalence settings. A negative PERC result does not exclude PE in moderate- or high-prevalence settings nor in moderate or high-risk patients.

Many hospitals employ the use of an age-adjusted d-dimer, which can complement the PERC rule.

PERC is a valuable aid to rule out PE only in patients who are pre-determined to be low risk by physician gestalt. When used properly, a PERC-negative result can confidently obviate further PE testing in those patients. This tool should augment, not supersede, your clinical decision-making, so always keep in mind the full clinical picture of the patient in front of you. Keep in mind its limitations (e.g. not for use in pregnancy or higher-risk cases).

Management

A d-dimer should be considered in the low-risk patient who is not PERC negative. If the d-dimer is negative and clinical gestalt determines a pre-test probability < 15%, then no further testing for PE is required. If the d-dimer is positive, CT-angiography (CTA) should be performed, or V/Q scan if contraindications to CTA. Consider using age-adjusted d-dimer in this protocol.

A patient who is low-risk via physician gestalt (<15%) and is PERC negative can have PE safely excluded from the differential.

Critical Actions

Patients at moderate-to-high risk for PE should not be evaluated using this tool. Patients who are not suspected of having PE do not need to have this rule applied. Patients who are low-risk for PE should be evaluated with this tool, as it can avoid further testing. Maintain a broad differential in patients with pleuritic complaints. Use caution in high-prevalence PE settings.

Evidence

The original 2004 PERC study was a prospective, multicenter trial with derivation and validation phases.

  • Derivation: 3,148 patients across 10 sites; 21 variables analyzed, 8 chosen for the final rule.

  • Validation arm: 1,427 low-risk and 382 very-low-risk patients from 2 sites.

    • Low-risk: sensitivity 96%, specificity 27%

    • Very low-risk: sensitivity 100%, specificity 15%

    • 90-day false negative rate: 1.4%, below the 1.8% testing threshold.

The 2008 multicenter validation expanded on this work and formally defined low pretest probability as <15%.

  • 8,138 patients from 13 sites (some overlapping with the 2004 cohort).

  • Physicians estimated pretest probability as <15%, 15–40%, or >40%.

  • 20% were deemed low-risk (<15%).

  • Among PERC-negative, low-risk patients:

    • 45-day false negative rate 1.0%, sensitivity 97.4%, specificity 21.9%.

Original Validation/Derivation study

Kline and colleagues conducted a prospective multicenter study evaluating clinical criteria designed to reduce unnecessary diagnostic testing in emergency department patients with suspected pulmonary embolism. The study, published in the Journal of Thrombosis and Haemostasis in 2004, enrolled 3,148 adult patients across 10 sites. Eligible patients were those presenting to the emergency department with sufficient clinical suspicion for pulmonary embolism that a board-certified emergency physician believed a formal PE evaluation was warranted. No exclusion criteria were applied.

The primary endpoint was clinically important venous thromboembolic disease, defined as treatment for pulmonary vascular thrombosis or definite deep venous thrombosis, or diagnosis of either entity at autopsy. Among patients categorized as low risk, the rule demonstrated a sensitivity of 96% and a specificity of 27%. In the very low-risk group, sensitivity was 100% and specificity was 15%. The false-negative rate at 90 days among low-risk patients was 1.4%. Based on these findings, the authors concluded that patients who meet none of the PERC criteria do not require D-dimer testing.

This study has several important strengths. It was prospective, multicenter, and performed in a real-world emergency department population, which improves its clinical relevance. It also derived a decision tool aimed at addressing a common and practical problem in emergency medicine, and its primary endpoint was clinically meaningful. One important limitation, however, is that one of the validation subsets consisted of a convenience sample of “very low-risk” patients, which may have introduced selection bias.

Additional validation study

Kline and colleagues also performed a prospective, non-interventional multicenter study validating the PERC rule in 8,138 emergency department outpatients undergoing testing for pulmonary embolism. Patients were included if clinicians ordered objective PE testing, such as D-dimer, V/Q scan, or CT angiography. Exclusion criteria included a known positive pulmonary vascular imaging study within the prior 7 days, inability to ensure follow-up within the same hospital system, or other concern for loss to follow-up. Funded by the NIH and the Emergency Medicine Foundation, the study found that among PERC-negative patients with a pretest probability below 15%, the 45-day false-negative rate was 1.0%, with sensitivity of 97.4% and specificity of 21.9%. Strengths included its large size and diverse multicenter design. Limitations were that low-risk status was determined by physician gestalt, which is subjective, and that the study was conducted mainly in U.S. emergency departments, which may limit generalizability.

Systematic Reviews/Meta-analyses

Several major guidelines support the use of the Pulmonary Embolism Rule-out Criteria in carefully selected low-risk patients. The 2018 ACEP clinical policy gives a Level B recommendation for using PERC to exclude pulmonary embolism in low-risk emergency department patients without further diagnostic testing. NICE similarly recommends PERC in low-risk patients, defined as those with a clinician-estimated PE probability below 15% when alternative diagnoses are also plausible. The American Society of Hematology also gives a conditional recommendation that a negative PERC can rule out PE in patients with low pretest probability, avoiding D-dimer testing or imaging. The 2019 PERT Consortium likewise endorses PERC in patients with low pretest probability. In contrast, the 2019 ESC/ERS guidelines do not recommend PERC in any patient group, citing insufficient evidence.

Summary of pertinent Clinical Practice Guidelines

ACEP Clinical Policy: Critical Issues in the Evaluation and Management of Adult Patients Presenting to the Emergency Department With Suspected Acute Venous Thromboembolic Disease 2018
https://www.acep.org/siteassets/new-pdfs/clinical-policies/clinical.policy.suspected.acute.venous.thromboembolic.disease.pdf		 Level B recommendations: For patients who are at low risk for acute PE, use PERC to exclude the diagnosis without further diagnostic testing.
National Institute for Health and Care Excellence (NICE, 2020) Use PERC in low-risk patients (clinician’s estimated PE likelihood <15%, with other diagnoses plausible).
European Society of Cardiology (ESC) & European Respiratory Society – ESC/ERS (2019) Do not recommend use of PERC in any patient group, citing lack of sufficient evidence.
American Society of Hematology (ASH, 2018) Conditional recommendation that in patients with a low pretest probability of PE, a negative PERC can be used to rule out PE without proceeding to D-dimer or imaging
PERT Consortium (2019 consensus) Use PERC (with a low pretest probability) to rule out PE

Formula

PE cannot be ruled out if any of the following are present

  • Age ≥ 50

  • HR ≥ 100

  • O₂ saturation on room air < 95%

  • Unilateral leg swelling

  • Hemoptysis

  • Recent trauma or surgery

  • Prior PE or DVT

  • Hormone use (oral contraceptives, hormone replacement, or estrogenic hormone use in male or female patients).

Facts & Figures

The PERC rule cannot be used to rule out PE in a patient with ANY of the above criteria.

Literature

Original derivation/validation studies

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):1247-1255.
doi:10.1111/j.1538-7836.2004.00790.x.

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

Appendix

Other validations

https://www.sciencedirect.com/science/article/pii/S0735675707003075
Assessment of the pulmonary embolism rule-out criteria rule for evaluation of suspected pulmonary embolism in the emergency department		 This was a secondary analysis in a Denver ED, with a small sample size (26 patients), high PE prevalence ~12% and wide confidence intervals. PERC showed 100% sensitivity (no missed PEs; CI up to 21% due to sample size) with 16% specificity in the overall cohort. Among the subset with low clinical probability, sensitivity was 100% and specificity 33% (NPV 100%).
https://www.jthjournal.org/article/S1538-7836%2822%2916509-8/fulltext
Application of pulmonary embolism rule-out criteria to the UK Manchester Investigation of Pulmonary Embolism Diagnosis (MIOPED) study cohort		 This was a retrospective analysis (published as a letter) applying the PERC rule to the existing MIOPED cohort in Manchester, UK. Although the exact sample size wasn’t specified, the analysis included low-risk patients with complete PERC data. They found a low false-negative rate (<2%), suggesting that PERC could safely exclude PE in this UK setting. As a brief, post hoc analysis without detailed methodology or performance metrics, its findings are limited by potential bias, lack of generalizability, and absence of full peer-reviewed results.
https://www.jthjournal.org/article/S1538-7836%2822%2907507-9/fulltext
The pulmonary embolism rule-out criteria (PERC) rule does not safely exclude pulmonary embolism		 This was a retrospective multicenter study evaluating the PERC rule in 1,675 patients across six European emergency departments, drawn from a higher-risk cohort with a 21% PE prevalence. Only 13.2% of patients were PERC-negative, and among them, 5.4% had PE. Another low-clinical probability subgroup had PE prevalence of 6.4%. The negative likelihood ratio was 0.21 overall but less favorable (0.63) in the low-risk subgroup. The study demonstrated that PERC performs poorly when applied to populations not preselected for very low PE risk, reinforcing that it should only be used after careful clinical risk stratification.
https://www.thelancet.com/journals/lanhae/article/PIIS2352-3026%2817%2930210-7/abstract
Pulmonary embolism rule-out criteria (PERC) rule in European patients with low implicit clinical probability (PERCEPIC): a multicentre, prospective, observational study		 The PERCEPIC study was a large prospective validation of the PERC rule in 1,757 patients across 12 French and Belgian EDs. Among patients deemed low-risk by clinician gestalt, 337 were PERC-negative, with only 4 diagnosed with VTE within 3 months. This represents a 1.2% miss rate (95% CI 0.4–2.9%) and no PE-related deaths. While the study confirmed safety, it was observational.
https://journals.lww.com/euro-emergencymed/abstract/2018/06000/retrospective_validation_of_the_pulmonary_embolism.6.aspx
Retrospective validation of the pulmonary embolism rule-out criteria rule in 'PE unlikely' patients with suspected pulmonary embolism		 This was a single-center, retrospective validation of the PERC rule in 986 suspected PE cases at a UK emergency department, focusing on 940 low-risk patients by Wells’ criteria. All underwent d-dimer testing, and PERC was applied post hoc. PERC missed 3 PE cases (sensitivity ~91.4%, NPV 99.1%) compared to only 1 missed by D-dimer (sensitivity 97.1%, NPV 99.8%). While PERC showed high negative predictive value, it was less sensitive than d-dimer, prompting the authors to caution against using PERC as a substitute for d-dimer in all low-risk patients.
https://onlinelibrary.wiley.com/doi/10.1111/acem.14744
Safety of the pulmonary embolism rule-out criteria rule: Findings from the Registro Informatizado de la Enfermedad TromboEmbolica venosa (RIETE) registry.		 This was a retrospective analysis using the RIETE VTE registry, examining 48,903 confirmed PE cases across 26 countries to assess how often patients with PE would have been PERC-negative. Only 346 cases (0.7%) met all eight PERC criteria, suggesting a very low failure rate and high sensitivity (~99.3%) of the PERC rule in practice. PERC-negative patients tended to have milder disease with fewer complications. Many missed cases had high-risk factors like recent pregnancy or cancer; conditions not included in the original PERC rule.

Randomized Control Trial

https://pmc.ncbi.nlm.nih.gov/articles/PMC5838786/
Effect of the Pulmonary Embolism Rule-Out Criteria on Subsequent Thromboembolic Events Among Low-Risk Emergency Department Patients: The PROPER Randomized Clinical Trial.		 The PROPER trial was the first cluster-randomized trial testing a PERC-based strategy in 1,916 low-risk patients across 14 French EDs. Sites alternated between using PERC alone (no d-dimer) and usual care (d-dimer followed by imaging if positive). The 3-month thromboembolic event rate was 0.1% in the PERC group vs. 0% in controls, meeting non-inferiority criteria. PERC use significantly reduced CT imaging (13% vs. 23%) and slightly shortened ED stays and admissions. Although limited by the very low-risk population (~2% PE prevalence), the study offered strong evidence that PERC, when applied appropriately, safely reduces unnecessary imaging.