Revised Geneva Score for Pulmonary Embolism
Estimates the clinical pre-test probability of pulmonary embolism using entirely objective criteria — no subjective clinician gestalt required. Includes both the original weighted version and the simplified (equal-weight) version, with 3-tier and 2-tier interpretation models.
Calculate Revised Geneva Score
Select the version you wish to use. The Original version uses variable-weighted points (range 0–25). The Simplified version assigns 1 point per criterion (range 0–9), with equivalent diagnostic performance. Both are displayed simultaneously regardless of the version selected.
The Revised Geneva Score uses entirely objective criteria — unlike the Wells PE score, it does not include a subjective “PE most likely diagnosis” assessment. This makes it more reproducible across clinicians and particularly useful in research settings, nursing triage, and protocols where inter-rater reliability is critical. Both the original and simplified versions have been validated in prospective studies.
Understanding the Revised Geneva Score
The original Geneva score was published by Wicki et al. in 2001, but it included variables requiring arterial blood gas analysis, limiting its practicality. Le Gal et al. published the Revised Geneva Score in 2006, replacing arterial blood gas variables with simpler clinical and historical criteria that can be assessed at the bedside without any laboratory tests. This version is the one used in current clinical practice.
In 2008, Klok et al. further derived the Simplified Revised Geneva Score, assigning a single point to each criterion regardless of its original weight. Validation studies demonstrated that the simplified version has equivalent diagnostic performance to the weighted original, making it easier to remember and apply rapidly. Both versions are endorsed by ESC 2019 guidelines as acceptable alternatives to the Wells PE score.
Original (Weighted) Scoring
Age >65 = +1
Previous DVT/PE = +3
Surgery/fracture ≤1 month = +2
Active malignancy = +2
Unilateral leg pain = +3
Leg vein palpation pain + oedema = +4
Haemoptysis = +2
HR 75–94 bpm = +3
HR ≥95 bpm = +5
Maximum: 25 points
Simplified Scoring
Age >65 = +1
Previous DVT/PE = +1
Surgery/fracture ≤1 month = +1
Active malignancy = +1
Unilateral leg pain = +1
Leg vein palpation pain + oedema = +1
Haemoptysis = +1
HR 75–94 bpm = +1
HR ≥95 bpm = +1
Maximum: 9 points
Geneva vs. Wells — Key difference: Every criterion in the Revised Geneva Score is objective and measurable. The Wells PE score includes the highly weighted subjective criterion “PE is #1 diagnosis or equally likely” (+3 points), which makes it dependent on the assessor’s experience and biases. The trade-off is that clinical gestalt (incorporated by Wells) may capture risk factors not formally included in any scoring system. In head-to-head studies, both scores perform similarly for risk stratification.
Score Interpretation & Risk Stratification
Original (Weighted) — 3-Tier & 2-Tier
| Score | 3-Tier Category | PE Prevalence | Recommended Action |
|---|---|---|---|
| 0–3 | Low | ~8% | D-dimer → if negative, PE excluded; if positive, CTPA |
| 4–10 | Moderate | ~29% | D-dimer → if negative, PE excluded; if positive, CTPA |
| ≥11 | High | ~64% | CTPA directly — D-dimer insufficient to exclude at this probability |
2-tier cut-off (original): PE Unlikely ≤5 (~10% prevalence) → D-dimer first. PE Likely ≥6 (~37% prevalence) → CTPA first.
Simplified — 3-Tier & 2-Tier
| Score | 3-Tier Category | PE Prevalence | Recommended Action |
|---|---|---|---|
| 0–1 | Low | ~7–9% | D-dimer → if negative, PE excluded; if positive, CTPA |
| 2–4 | Moderate | ~28–30% | D-dimer → if negative, PE excluded; if positive, CTPA |
| ≥5 | High | ~64% | CTPA directly — D-dimer insufficient to exclude |
2-tier cut-off (simplified): PE Unlikely ≤2 → D-dimer first. PE Likely ≥3 → CTPA first.
Regardless of which version or tier model you use, the management principle is the same: low-to-moderate probability patients can have PE safely excluded with a negative D-dimer (use age-adjusted thresholds for patients >50), while high-probability patients require definitive imaging (CTPA or V/Q scan) regardless of D-dimer results. This is identical to the pathway used with the Wells PE score.
Geneva Criteria in Clinical Context
Each Geneva criterion was selected based on its independent association with PE in multivariate analysis. Understanding the clinical rationale for each variable helps ensure accurate scoring and meaningful interpretation.
Age is the only criterion that receives the same weight (+1) in both the original and simplified versions. The incidence of PE increases substantially with age — from approximately 1 per 1,000 person-years before age 50 to 6–7 per 1,000 person-years by age 80. Older patients also have higher rates of comorbid conditions (malignancy, immobility, heart failure) that contribute to VTE risk. The lower weight (+1) reflects the fact that age alone, without other risk factors, confers only a modest increase in pre-test probability.
Importantly, D-dimer levels also rise with age, which is why age-adjusted D-dimer thresholds (age × 10 µg/L for patients >50) are recommended when using D-dimer to exclude PE in elderly patients after Geneva scoring.
A history of objectively confirmed VTE is one of the strongest risk factors for recurrence. Annual recurrence rates after a first unprovoked VTE are approximately 5–10%, and higher in patients with persistent risk factors (thrombophilia, malignancy, immobility). The Geneva score weights this criterion heavily (+3 in the original) because a patient with prior VTE and new cardiopulmonary symptoms has a substantially higher pre-test probability of PE than the general population.
This criterion requires objectively confirmed prior VTE — not merely suspected or empirically treated thrombosis. If the previous episode was not confirmed by imaging, the clinician should use judgement about whether to score this criterion.
Surgery or fracture within 1 month (+2): Post-operative immobility, tissue factor release, and endothelial damage create the classic Virchow’s triad. Major orthopaedic surgery (hip/knee replacement), abdominal/pelvic surgery, and neurosurgery carry the highest VTE risk. The 1-month window captures the period of maximal risk, though some risk persists for up to 3 months after major procedures.
Active malignancy (+2): Cancer-associated thrombosis accounts for approximately 20% of all VTE. Mucin-secreting adenocarcinomas (pancreatic, ovarian, gastric, lung), brain tumours, and haematological malignancies carry particularly high risk. Active malignancy includes cancer diagnosed or treated within the preceding year, or cancer currently receiving treatment (chemotherapy, radiotherapy, surgery, or palliative care). Cancer in complete remission beyond 1 year does not score.
Unilateral lower limb pain (+3): Spontaneous leg pain — not provoked by recent trauma or exertion — suggests concurrent DVT. This is distinct from the next criterion (physical examination findings) and refers to the patient’s symptom, not the examination finding. The pain may be in the calf, thigh, or groin and is typically described as aching, heaviness, or cramping.
Pain on deep vein palpation + unilateral oedema (+4): This is the highest-weighted criterion in the original Geneva score. It combines a physical examination finding (tenderness on palpation along the deep venous system) with an observable sign (asymmetric leg oedema). Together, these findings strongly suggest DVT and therefore significantly increase the probability of PE. Note that both findings must be present — leg tenderness alone without oedema, or oedema alone without tenderness, should not score +4.
The Geneva score uniquely uses two heart rate thresholds rather than a single cut-off. A heart rate of 75–94 bpm scores +3, while a heart rate ≥95 bpm scores +5 in the original version. This graded approach reflects the dose-response relationship between tachycardia severity and PE probability — mild tachycardia may accompany smaller or more peripheral PE, while significant tachycardia suggests greater haemodynamic compromise.
In the simplified version, both thresholds score +1. This means the simplified score loses some discriminating power from heart rate, but validation studies show that overall diagnostic performance is maintained. The heart rate should be measured at the initial clinical assessment — not after anxiolytic treatment or during a pain-free period, as the presenting vital signs are what the score was validated against.
Haemoptysis is present in fewer than 10% of confirmed PE cases but is a relatively specific indicator of pulmonary infarction. PE-related haemoptysis is typically small-volume (blood-streaked sputum rather than massive haemoptysis) and accompanies pleuritic chest pain. Massive haemoptysis (>200 mL in 24 hours) is more suggestive of other aetiologies (bronchiectasis, lung cancer, aspergilloma, tuberculosis) and warrants a different diagnostic approach.
Any amount of haemoptysis — even a single episode of blood-tinged sputum — should be scored. The criterion does not differentiate by volume or duration. In the setting of dyspnoea and pleuritic chest pain, haemoptysis substantially raises the pre-test probability of PE and should lower the threshold for imaging.
Use Geneva when: You want purely objective, reproducible scoring — especially useful in triage protocols, nursing-initiated pathways, research settings, or when multiple clinicians of varying experience may be involved. Also preferred when there is concern about subjective bias in the “PE most likely” criterion of Wells.
Use Wells when: The clinician’s gestalt adds meaningful information (e.g., experienced emergency physicians who can integrate pattern recognition). Wells is also more widely known and may be preferred in settings with established Wells-based institutional protocols. Both scores are accepted by ESC, ACEP, and BTS guidelines — there is no “wrong” choice.
Special Populations & Considerations
The Revised Geneva Score was derived and validated in emergency department populations in Europe. Its performance may differ in certain clinical scenarios.
PERC rule and Geneva: The PERC rule was validated primarily with the Wells PE score, not the Geneva score. However, several studies have applied PERC to patients categorised as low-risk by Geneva with similar results. If using Geneva in a low-probability context (score 0–3 on original, or 0–1 on simplified), PERC may reasonably be considered, though this is an off-label application and should be documented as such.
Diagnostic Pathway Using the Geneva Score
The diagnostic approach using the Geneva score mirrors the pathway used with Wells. The key decision points — when to use D-dimer, when to proceed to imaging, and when imaging alone is sufficient — are the same regardless of which clinical prediction rule is used.
Before applying any clinical decision rule, assess haemodynamic stability. Patients with sustained systolic BP <90 mmHg, obstructive shock, or cardiac arrest with suspected PE require immediate CTPA or bedside echocardiography — clinical prediction rules are irrelevant in this context. Consider empiric thrombolysis for high-risk (massive) PE with haemodynamic collapse.
For haemodynamically stable patients, calculate the Revised Geneva Score using either the original or simplified version. Classify the result using the 3-tier model (Low / Moderate / High) or the 2-tier model (PE Unlikely / PE Likely). The 3-tier model provides more granular risk stratification, while the 2-tier model simplifies the pathway into a single binary decision.
Document the version used (original or simplified), the individual criteria scored, the total, and the resulting risk category. This is essential for medicolegal defensibility and continuity of care.
For patients with low or moderate probability (original 0–10 or simplified 0–4), or “PE Unlikely” on the 2-tier model, proceed to high-sensitivity D-dimer testing. Use age-adjusted thresholds (age × 10 µg/L) for patients over 50 years.
Negative D-dimer: PE is safely excluded (3-month VTE rate <1%). No imaging needed. Positive D-dimer: Proceed to CTPA or V/Q scan. A positive D-dimer does not diagnose PE — it indicates that imaging is required. Consider alternative causes of elevated D-dimer (infection, malignancy, surgery, age, pregnancy).
For patients with high probability (original ≥11 or simplified ≥5), or “PE Likely” on the 2-tier model, proceed directly to CTPA. D-dimer is insufficient to exclude PE at this pre-test probability — the false-negative rate is clinically unacceptable. V/Q scintigraphy is an alternative when CTPA is contraindicated.
If CTPA is positive, risk-stratify (haemodynamics, RV function, troponin, BNP) and initiate anticoagulation. If CTPA is negative despite high clinical probability, consider lower limb compression ultrasound, V/Q scan, or specialist review — a negative CTPA in the context of high pre-test probability warrants further evaluation.
Common Pitfalls & Limitations
The Revised Geneva Score is well-validated but has specific limitations and common sources of misapplication that clinicians should be aware of.
Three distinct versions of the Geneva score exist in the literature, causing frequent confusion: the Original Geneva Score (Wicki 2001, required ABG), the Revised Geneva Score (Le Gal 2006, clinical criteria only), and the Simplified Revised Geneva Score (Klok 2008, each criterion = 1 point). Each version has different score ranges and different risk-category thresholds. Applying the wrong thresholds to the wrong version can lead to misclassification.
Current guidelines refer to the Revised Geneva Score (2006) and its simplified variant (2008). The original 2001 score is considered obsolete. When documenting a Geneva score, always specify which version was used and the corresponding risk-category thresholds.
The heart rate criteria (which contribute up to +5 points in the original score) can be significantly influenced by medications and comorbidities. Beta-blockers, calcium channel blockers (verapamil, diltiazem), and digoxin blunt the tachycardic response to PE, potentially causing underscoring. Conversely, pain, anxiety, fever, anaemia, thyrotoxicosis, and stimulant use can elevate heart rate in the absence of PE, causing overscoring.
Clinicians should be aware that the Geneva score may underestimate PE probability in patients on rate-controlling medications and overestimate it in patients with non-PE causes of tachycardia. This is an inherent limitation of using objective vital sign criteria — they sacrifice the nuance of clinical interpretation for reproducibility.
The Geneva score’s greatest strength — pure objectivity — is also its limitation. By excluding clinician gestalt, the score cannot capture atypical presentations, unusual risk factor combinations, or subtle clinical patterns that an experienced emergency physician might recognise. A patient who “looks like PE” to an experienced clinician but scores low on Geneva may be at higher risk than the score suggests.
This limitation is most relevant for experienced clinicians in well-resourced settings. The Geneva score is most valuable where inter-rater reliability matters: triage protocols, nurse-initiated pathways, junior doctor decision support, and research settings. Where clinical experience is available and reliable, the Wells PE score (which includes gestalt) may provide incremental value — though both perform similarly in validation studies.
The Geneva score has two separate DVT-related criteria that are commonly confused: “unilateral lower limb pain” (+3 in original) refers to the patient’s symptom, while “pain on deep vein palpation + unilateral oedema” (+4 in original) refers to an examination finding. These are scored independently — a patient who has both spontaneous leg pain and examination-positive DVT signs would score +3 and +4 (total +7) on these criteria alone.
However, the examination criterion requires both components: tenderness on palpation of deep veins and unilateral oedema. Tenderness without oedema, or oedema without tenderness, does not qualify for the +4 points. This distinction matters because the examination criterion carries the highest individual weight in the original score.
As with the Wells score, using D-dimer alone to exclude PE in high-probability patients (Geneva ≥11 original, or ≥5 simplified) is a patient safety error. At a PE prevalence of approximately 64%, the negative predictive value of D-dimer is insufficient to safely exclude PE. These patients require definitive imaging (CTPA or V/Q scan) regardless of D-dimer results.
This error is more likely when D-dimer is ordered reflexively before the Geneva score is calculated. The score should always be determined first, so that subsequent test results are interpreted in the correct probabilistic framework. When the Geneva score indicates high probability, a negative D-dimer should not override the clinical assessment — proceed to imaging.
Quick Reference Summary
| Category | Original Score | Simplified Score | Action |
|---|---|---|---|
| Low | 0–3 | 0–1 | D-dimer (age-adjusted if >50) → negative excludes PE |
| Moderate | 4–10 | 2–4 | D-dimer → negative excludes PE; positive → CTPA |
| High | ≥11 | ≥5 | CTPA directly — D-dimer insufficient to exclude |
The Golden Rule: The Geneva score is a fully objective PE probability tool — no subjective judgement required. Calculate the score → determine risk category → follow the D-dimer/CTPA pathway. Low or moderate? D-dimer decides. High? Image directly. This is the same pathway as Wells, just with an objective scoring system. Use age-adjusted D-dimer thresholds in patients over 50.
Disclaimer & References
For Educational Purposes Only. This calculator and the accompanying clinical information are intended as educational tools for healthcare professionals. They do not replace clinical judgement. Results should be interpreted in the full clinical context. Lab reference ranges vary by institution — verify with your own laboratory. Drug dosages should be confirmed against current prescribing information.
References
- Le Gal G, Righini M, Roy PM, et al. Prediction of pulmonary embolism in the emergency department: the revised Geneva score. Ann Intern Med. 2006;144(3):165–171. DOI: 10.7326/0003-4819-144-3-200602070-00004
- Klok FA, Mos ICM, Nijkeuter M, et al. Simplification of the revised Geneva score for assessing clinical probability of pulmonary embolism. Arch Intern Med. 2008;168(19):2131–2136. DOI: 10.1001/archinte.168.19.2131
- Wicki J, Perneger TV, Junod AF, Bounameaux H, Perrier A. Assessing clinical probability of pulmonary embolism in the emergency ward: a simple score. Arch Intern Med. 2001;161(1):92–97. DOI: 10.1001/archinte.161.1.92
- Konstantinides SV, Meyer G, Becattini C, et al. 2019 ESC Guidelines for the diagnosis and management of acute pulmonary embolism. Eur Heart J. 2020;41(4):543–603. DOI: 10.1093/eurheartj/ehz405
- 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):1117–1124. DOI: 10.1001/jama.2014.2135
- Ceriani E, Combescure C, Le Gal G, et al. Clinical prediction rules for pulmonary embolism: a systematic review and meta-analysis. J Thromb Haemost. 2010;8(5):957–970. DOI: 10.1111/j.1538-7836.2010.03801.x
- Douma RA, Mos ICM, Erkens PMG, et al. Performance of 4 clinical decision rules in the diagnostic management of acute pulmonary embolism: a prospective cohort study. Ann Intern Med. 2011;154(11):709–718. DOI: 10.7326/0003-4819-154-11-201106070-00002
- Kearon C, de Wit K, Parpia S, et al. Diagnosis of pulmonary embolism with D-dimer adjusted to clinical probability. N Engl J Med. 2019;381(22):2125–2134. DOI: 10.1056/NEJMoa1909159