Pulmonary embolism
Description
- Occlusion of pulmonary arteries, almost always by thrombus embolised from proximal lower-limb or pelvic DVT
- One end of the VTE spectrum - DVT and PE are the same disease
Classification - 2026 AHA/ACC clinical categories
*"Massive" and "submassive" are retired.* Five categories, graded by clinical severity, biomarkers, RV function and haemodynamics.
| Category | Definition |
|---|---|
| A | Asymptomatic/subclinical - incidental on CT for another reason |
| B | Symptomatic, low severity: PESI I-II, sPESI 0, or Hestia 0 |
| C1 | Elevated severity score (PESI III-V, sPESI >=1, Hestia >=1); normal troponin, no RV dysfunction |
| C2 | Elevated score + EITHER raised troponin OR RV enlargement/dysfunction |
| C3 | Elevated score + BOTH raised troponin AND RV dysfunction |
| D1 | Transient or recurrent hypotension - short-lived or fluid-responsive, no end-organ hypoperfusion |
| D2 | Normotensive shock - normal BP but raised lactate / AKI / hypoperfusion |
| E1 | Persistent hypotension with cardiogenic shock |
| E2 | Refractory cardiogenic shock (SCAI D-E) or cardiac arrest without ROSC after 30 min |
- Respiratory modifier "R" appended (e.g. C3R, D2R) when hypoxaemia, tachypnoea or supplemental oxygen requirement is present
- The old "submassive" bucket spanned everything from a mildly raised troponin to imminent collapse - C1 to D2 is the point of the new scheme
- ESC 2019 equivalents still widely used in Australia: low / intermediate-low / intermediate-high / high risk
Special forms
- Saddle PE - straddles the bifurcation; anatomical, not physiological - many are haemodynamically stable
- Subsegmental PE - often incidental; risk-benefit of anticoagulation genuinely uncertain in isolated subsegmental PE without DVT
- Non-thrombotic - fat (long bone fracture, 24-72 h, petechiae + confusion + hypoxia), amniotic fluid, air, tumour, septic (right-sided endocarditis, IVDU), talc, hydatid
Epidemiology
- VTE incidence ~1-2/1000/yr; ~17,000 Australians/yr, ~5,000 deaths
- PE is the commonest preventable cause of hospital death
- Incidence rises steeply with age (~1% per year over 80)
- ~2/3 of VTE is hospital-associated (during or within 90 days of admission)
- ~50% of proximal DVT have asymptomatic PE; ~70% of PE have demonstrable DVT
- Untreated PE mortality ~25-30%; treated ~2-8%
Aetiopathogenesis
Virchow's triad
- Stasis - immobility, surgery, long-haul travel, plaster, paralysis, hospitalisation
- Endothelial injury - surgery, trauma, central lines, prior VTE
- Hypercoagulability - inherited or acquired
Provoking factors - stratify by strength (determines duration of anticoagulation)
| Strength | Factors |
|---|---|
| Major transient | Surgery >30 min under GA, major trauma, fracture, hospitalisation >=3 days with immobility, caesarean (all within 3 months) |
| Minor transient | Minor surgery, oestrogen (COCP, HRT), pregnancy/puerperium, leg injury with reduced mobility, long-haul flight >8 h, acute medical admission |
| Persistent | Active cancer, antiphospholipid syndrome, IBD, chronic inflammatory disease, nephrotic syndrome, obesity, paralysis, indwelling catheter |
| Unprovoked | None identified - the highest recurrence risk |
Thrombophilia
- Inherited: factor V Leiden (~5% of Caucasians, commonest), prothrombin G20210A, protein C/S deficiency, antithrombin deficiency (most thrombogenic)
- Acquired: antiphospholipid syndrome, malignancy, PNH, myeloproliferative neoplasm (JAK2 - especially splanchnic vein thrombosis), HIT, nephrotic syndrome
- *Testing rarely changes management* - it does not alter initial treatment or usually duration
- Test if: APS suspected (arterial + venous, recurrent miscarriage, livedo, thrombocytopenia), unusual site, strong family history in a young patient, or where the result would change a decision
- Never test during acute thrombosis or on anticoagulation - protein C/S and antithrombin are falsely low, lupus anticoagulant unreliable on DOAC
Pathophysiology - what kills
- Mechanical obstruction + hypoxic vasoconstriction + vasoactive mediators (serotonin, thromboxane)
- -> abrupt inc pulmonary vascular resistance
- RV afterload rises acutely against a thin-walled ventricle it cannot overcome
- -> RV dilatation -> interventricular septal bowing -> impaired LV filling -> dec cardiac output -> hypotension
- -> dec coronary perfusion to a dilated, hypertensive RV -> RV ischaemia -> further failure
- *This spiral, not hypoxaemia, is the mechanism of death in PE*
- Gas exchange: inc dead space, V/Q mismatch, right-to-left shunt (PFO), atelectasis from surfactant loss
- -> hypoxaemia + hyperventilation -> respiratory alkalosis with a widened A-a gradient
- Pulmonary infarction occurs in only ~10% (dual blood supply) - typically peripheral, small emboli, causing pleuritic pain and haemoptysis
- A large central PE causes shock without pleuritic pain; a small peripheral one causes pleuritic pain without shock
Diagnosis
Presentation
- Dyspnoea (most common), then pleuritic chest pain, tachypnoea, tachycardia
- Haemoptysis, syncope (implies significant haemodynamic effect), presyncope, anxiety, cough
- Signs of DVT in ~1/3; low-grade fever; pleural rub; inc JVP, parasternal heave, loud P2 if RV strain
- *Presentation is non-specific - the diagnosis is made by thinking of it*
Structured probability assessment first
Wells score (PE)
| Feature | Points |
|---|---|
| Clinical signs of DVT | 3 |
| PE is the most likely diagnosis | 3 |
| HR >100 | 1.5 |
| Immobilisation >=3 days or surgery in 4 weeks | 1.5 |
| Previous DVT/PE | 1.5 |
| Haemoptysis | 1 |
| Malignancy | 1 |
- <=4 "PE unlikely" -> D-dimer; >4 "PE likely" -> straight to CTPA
- PERC rule - in a genuinely low-probability patient, if all 8 negative (age <50, HR <100, SpO2 >=95%, no haemoptysis, no oestrogen, no prior VTE, no unilateral leg swelling, no surgery/trauma in 4 weeks), PE is excluded without D-dimer
D-dimer
- High sensitivity, poor specificity - a rule-out test only
- Age-adjusted cutoff: age x 10 microg/L (FEU) if age >50 - improves specificity without losing sensitivity
- Falsely raised: age, pregnancy, cancer, infection, surgery, trauma, inflammation, AF, haemodialysis
- Do not order in a high-probability patient - a negative result will not stop you imaging
CTPA - the standard
- Sensitivity and specificity >95% for segmental or larger
- Limitations: contrast nephropathy, iodine allergy, radiation (breast dose in young women), over-diagnosis of subsegmental emboli
- V/Q or V/Q SPECT if contrast contraindicated, renal failure, pregnancy (with a normal CXR), or young women
- Bilateral leg compression ultrasound - a proximal DVT in a patient with symptoms suggesting PE is enough to treat
Supporting tests - useful for risk, not diagnosis
- ECG: sinus tachycardia is the commonest finding; S1Q3T3 is classic but present in <20%; RBBB, right axis, T-wave inversion V1-V4 (RV strain - correlates with severity), AF
- CXR: usually normal; Hampton hump (peripheral wedge), Westermark sign (regional oligaemia), raised hemidiaphragm, small effusion
- ABG: hypoxaemia, hypocapnia, widened A-a gradient; a normal ABG does not exclude PE
- Troponin and BNP/NT-proBNP - RV strain; prognostic, and required for the C1/C2/C3 split
- Echocardiography - RV dilatation (RV:LV >1), hypokinesis with apical sparing (McConnell sign), 60/60 sign, septal flattening, TR jet, rarely thrombus in transit
- In shock with suspected PE and no time for CT, RV dilatation on bedside echo justifies thrombolysis
PE in pregnancy
- Start with leg ultrasound; if positive, treat and stop
- CXR first; if normal -> V/Q or perfusion-only scan; if abnormal -> CTPA
- Both modalities are acceptable; the fetal dose from either is far below the threshold for harm. The risk of missing PE greatly exceeds the radiation risk**
- D-dimer is unreliable; YEARS-adapted and pregnancy-adapted algorithms are validated
Risk stratification tools
- PESI / sPESI - clinical only; sPESI 0 identifies candidates for outpatient care
- sPESI: age >80, cancer, chronic cardiopulmonary disease, HR >=110, SBP <100, SpO2 <90%
- Hestia criteria - a checklist for outpatient eligibility (haemodynamics, oxygen need, bleeding risk, renal/hepatic failure, pregnancy, social)
- Troponin + RV assessment - stratify intermediate risk
- Lactate - identifies normotensive shock (Category D2)
Management
1. Immediate
- Oxygen, IV access, monitoring; PE response team (PERT) activation if intermediate-high or higher
- *Start therapeutic anticoagulation on clinical suspicion while awaiting imaging* when probability is intermediate/high and bleeding risk is acceptable
- Fluids cautiously - 250-500 mL only. Volume loading a failing RV worsens septal bowing and can precipitate arrest
- Noradrenaline first-line vasopressor; add dobutamine for RV inotropy
- Avoid intubation if at all possible - induction and positive pressure ventilation can cause cardiovascular collapse. If unavoidable: ketamine, avoid high PEEP, keep preload
2. Anticoagulation
| Situation | Agent |
|---|---|
| Most patients | DOAC - apixaban or rivaroxaban (no lead-in heparin needed); dabigatran/edoxaban need >=5 days parenteral lead-in |
| Haemodynamic instability, likely reperfusion, high bleeding risk | IV unfractionated heparin (short half-life, reversible) |
| Severe renal impairment (CrCl <15-30) | UFH, then warfarin |
| Antiphospholipid syndrome (triple positive) | Warfarin - DOACs are inferior (TRAPS) |
| Cancer-associated | DOAC (apixaban preferred) or LMWH; avoid rivaroxaban/edoxaban with luminal GI or GU tumours - inc bleeding |
| Pregnancy | LMWH - DOACs and warfarin are contraindicated |
| Mechanical valve | Warfarin |
- Apixaban 10 mg BD x 7 days -> 5 mg BD; rivaroxaban 15 mg BD x 21 days -> 20 mg daily
- Thrombocytopenia/HIT -> argatroban or fondaparinux; never heparin
3. Reperfusion - by category
| Category | Approach |
|---|---|
| A | Anticoagulate; can generally be discharged from ED |
| B | Anticoagulate; outpatient or early discharge if sPESI 0/Hestia negative and social circumstances allow |
| C1 | Anticoagulate, admit, monitor |
| C2-C3 | Anticoagulate, monitor closely for deterioration; consider PERT and advanced therapy if deteriorating |
| D1-D2 | Advanced therapy can be considered - catheter-directed thrombolysis, mechanical thrombectomy, systemic lysis |
| E1-E2 | Reperfusion is reasonable/indicated - systemic thrombolysis, catheter-directed lysis, mechanical thrombectomy, or surgical embolectomy |
- Systemic thrombolysis: alteplase 100 mg over 2 h (or 0.6 mg/kg over 15 min in arrest)
- Reduces haemodynamic decompensation and mortality in high-risk PE
- *Major bleeding ~10%, intracranial haemorrhage ~2-3%*
- PEITHO: in intermediate-high risk, lysis prevented decompensation but did not reduce mortality and caused excess ICH -> do not thrombolyse routine intermediate-risk PE
- Catheter-directed thrombolysis / large-bore mechanical thrombectomy - lower lytic dose or none; PEERLESS compared the two in intermediate-risk disease
- Growing use; the 2026 guideline is more permissive than ESC 2019, which reserved these for failed or contraindicated systemic lysis
- VA-ECMO - bridge in E2 (refractory shock or arrest)
- Surgical embolectomy - contraindication to lysis, failed lysis, or thrombus in transit/PFO
- IVC filter - only for absolute contraindication to anticoagulation or recurrence despite therapeutic anticoagulation. Retrieve as soon as anticoagulation can start
4. Duration of anticoagulation - by provoking factor
| Setting | Duration |
|---|---|
| Provoked by a major transient risk factor | 3 months, then stop |
| Provoked by a minor transient risk factor | 3-6 months, individualise |
| Unprovoked | >=3 months, then consider indefinite - recurrence ~10% at 1 yr, ~30% at 5-10 yr |
| Active cancer | While cancer is active or on treatment |
| Antiphospholipid syndrome | Indefinite warfarin |
| Recurrent unprovoked VTE | Indefinite |
- Extended-phase reduced-dose DOAC - apixaban 2.5 mg BD or rivaroxaban 10 mg daily after the first 6 months (EINSTEIN-CHOICE, AMPLIFY-EXT) - similar efficacy, less bleeding
- Weigh HAS-BLED / bleeding risk, patient preference, D-dimer after cessation, male sex (higher recurrence)
- Aspirin is markedly inferior to anticoagulation for secondary prevention
5. Cancer screening after unprovoked VTE
- History, examination, basic bloods, CXR, and age-appropriate screening only
- *Extensive occult-cancer screening (CT abdomen/pelvis, tumour markers) does not improve outcomes (SOME trial)*
6. Prevention
- Risk-assess every hospital inpatient; mechanical + pharmacological prophylaxis unless contraindicated
- Extended prophylaxis after major orthopaedic and abdominopelvic cancer surgery
7. Follow-up
- Review at 3-6 months for persistent dyspnoea -> screen for CTEPH with echo, then V/Q scan (the screening test of choice - CTPA misses chronic disease)
- Reassess bleeding risk and duration decision at each review
- Post-PE syndrome: deconditioning, anxiety - pulmonary rehabilitation helps
Associations
- DVT - the same disease; look for it
- Malignancy - ~5-10% of unprovoked VTE have occult cancer; pancreas, lung, ovary, brain, stomach, myeloproliferative
- Antiphospholipid syndrome - arterial + venous thrombosis, recurrent fetal loss, livedo reticularis, thrombocytopenia
- Inherited thrombophilia - factor V Leiden, prothrombin gene, protein C/S, antithrombin
- Pregnancy and puerperium - leading direct cause of maternal death
- Oestrogen - COCP, HRT, tamoxifen; inc risk with factor V Leiden (multiplicative)
- Nephrotic syndrome, IBD, PNH, myeloproliferative neoplasms (JAK2)
- Heparin-induced thrombocytopenia, vaccine-induced immune thrombotic thrombocytopenia
- COVID-19 and other severe infection
- CTEPH - the chronic complication
- PFO - paradoxical embolism -> stroke; also worsens hypoxaemia acutely
- Obesity, smoking, immobility, long-haul travel, IV drug use
Natural history & complications
- Most deaths occur within the first hours - before or shortly after diagnosis
- Mortality by category: A-B <1%; C1-C3 ~3-15%; D ~15-25%; E >30-50%
- Anticoagulation reduces mortality from ~25-30% to ~2-8%
- ~10% of PE deaths occur in patients diagnosed and treated - often from RV failure in the first 48 h
Recurrence
- Provoked by a major transient factor: ~1-3%/yr after stopping
- Unprovoked: ~10% at 1 yr, ~30% at 5-10 yr
- Higher in: male sex, obesity, persistent risk factor, raised D-dimer after cessation, residual thrombus, antiphospholipid syndrome
- Recurrence tends to replicate the index event - a PE tends to recur as a PE
Complications
- RV failure and cardiogenic shock - the acute killer
- CTEPH in ~2-4% at 2 years
- Persistent dyspnoea beyond 3 months -> echo -> V/Q scan
- Pulmonary endarterectomy is potentially curative - refer every case to a specialist centre before assuming it is inoperable
- Balloon pulmonary angioplasty and riociguat for inoperable/residual disease
- Post-PE syndrome - exertional limitation, deconditioning, anxiety, reduced QoL in up to 50%
- Bleeding on anticoagulation - major bleeding ~2-3%/yr
- Post-thrombotic syndrome of the leg (from the DVT)
- Pulmonary infarction, effusion, secondary infection
Monitor
- Symptoms of recurrence and of bleeding; renal function (DOAC dosing); adherence
- Dyspnoea at 3-6 months -> CTEPH screening
- Review duration decision at every visit
🔒
6 more sections, plus exam facts
Premium unlocks every note across every specialty, and the full exam fact library behind it.
Get premium access