Pulmonology
Pulmonary Embolism
When a leg clot lodges in the lung — diagnosis, hemodynamics, and treatment
Pulmonary embolism is the obstruction of a pulmonary artery branch by a clot — almost always traveling from a deep vein thrombosis in the legs. Sudden dyspnea, hypoxia, and right heart strain. CTPA confirms, anticoagulation treats, and thrombolytics are reserved for hemodynamic collapse.
- Hospital incidence1-2% of all admissions
- Mortality untreated~30%
- Mortality treated2-8%
- Gold-standard imagingCT pulmonary angiography
- Source of clotDVT in >90% (proximal leg veins)
- First-line treatmentDOAC (apixaban, rivaroxaban)
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
How a leg clot becomes a lung clot
A clot in a calf vein is, in itself, not immediately dangerous to the lungs. The danger begins when it extends — propagates — into the popliteal, femoral, or iliac veins, where the lumen is wide, the flow is laminar, and a fragment can detach and travel free. From there the path is mechanically inevitable: the embolus rides venous flow upward through the inferior vena cava, into the right atrium, across the tricuspid valve into the right ventricle, and out through the pulmonary trunk. The first vessel that is narrower than the embolus is where it stops.
What it does on arrival depends on size and pre-existing reserve. A small subsegmental clot lodges peripherally, causes a wedge-shaped infarction (if collateral bronchial circulation is also impaired), and may produce only a mild rise in heart rate and a transient pleuritic ache. A massive central clot — a "saddle embolus" straddling the bifurcation of the main pulmonary artery — obstructs both lungs simultaneously, the right ventricle dilates within minutes, cardiac output collapses, and the patient may die before the diagnosis is even considered.
V/Q mismatch and hypoxia
A lung segment without perfusion is the textbook example of dead space: ventilation continues, but no blood reaches the alveoli to pick up oxygen or release CO₂. This in itself does not cause hypoxia — you can hyperventilate normal lung to compensate for the CO₂. The hypoxia of PE comes from elsewhere: collateral redistribution of blood floods the remaining circulation, lowering the V/Q ratio in still-perfused lung; surfactant breaks down in the unperfused segment, causing atelectasis; in massive PE, right-to-left shunting through a patent foramen ovale (present in 25% of adults) bypasses the lungs entirely; and severe hypotension reduces mixed venous oxygen content. The result: hypoxia disproportionate to the radiographic findings, often with a low PaCO₂ from compensatory hyperventilation.
Worked clinical example
A 62-year-old woman, three days post-total knee arthroplasty, develops sudden shortness of breath while walking to the bathroom. Heart rate 118, blood pressure 132/76, respiratory rate 26, SpO₂ 91% on room air. Right calf is mildly swollen and warm. She is on prophylactic enoxaparin 40 mg daily — appropriate but not therapeutic.
Wells score: signs of DVT (3) + recent surgery (1.5) + heart rate >100 (1.5) + PE more likely than alternative (3) = 9. High probability. No need for D-dimer (would not change management). She goes directly to CTPA, which shows bilateral lobar emboli with mild right ventricular dilation (RV:LV ratio 1.1, mildly elevated). Troponin slightly positive at 0.08 ng/mL, BNP 380. She is hemodynamically stable but at intermediate risk — the "submassive" category.
Treatment: full-dose anticoagulation. Apixaban 10 mg twice daily for 7 days, then 5 mg twice daily. Given the post-surgical setting and stable hemodynamics, no thrombolytic. Lower-extremity ultrasound confirms a popliteal DVT. Repeat echo at 24 hours shows normalizing RV function. She is discharged on day 4 with apixaban for 3 months (provoked event), counseled about prevention of post-thrombotic syndrome and a return-precaution plan.
Massive vs submassive vs low-risk PE
| Category | Hemodynamics | RV strain markers | 30-day mortality | Treatment |
|---|---|---|---|---|
| Massive (high-risk) | SBP <90 mmHg or vasopressors | Severe RV dysfunction | 15-65% | Systemic tPA, consider thrombectomy/ECMO |
| Submassive (intermediate-high) | Normotensive | RV dilation + elevated troponin/BNP | 3-15% | Anticoagulation; consider catheter-directed thrombolysis |
| Submassive (intermediate-low) | Normotensive | RV dilation OR cardiac biomarker (not both) | 2-5% | Anticoagulation; monitored unit |
| Low-risk | Normotensive | No RV strain, no biomarker elevation | <2% | Anticoagulation; consider outpatient management (sPESI ≤ 1) |
| Saddle embolus (anatomy, not severity) | Variable | Often severe | Variable | Stratify by hemodynamics, not the anatomy alone |
| Subsegmental PE | Normotensive | None | <1% | Anticoagulation in most; surveillance only in select low-risk cases |
Why PE matters
- Mortality math. PE causes ~100,000 US deaths annually, comparable to motor vehicle accidents.
- Hospital prophylaxis. Universal VTE risk assessment and mechanical/pharmacologic prophylaxis prevent thousands of in-hospital deaths each year.
- Cancer. Malignancy increases VTE risk 4-7×; some PEs are the presenting feature of occult cancer (Trousseau syndrome).
- Pregnancy. 5× increased risk; LMWH because warfarin and DOACs cross the placenta.
- Long-haul travel. Risk small but real; hydration, ambulation, and compression stockings for high-risk travelers.
- Post-PE survivorship. Up to 4% develop chronic thromboembolic pulmonary hypertension (CTEPH) — surgically curable with pulmonary thromboendarterectomy.
- Bleeding management. Andexanet for factor Xa inhibitors; idarucizumab for dabigatran; vitamin K and PCC for warfarin.
Common misconceptions
- "Normal pulse ox rules out PE." Up to 25% of PE patients have SpO₂ ≥95% on room air.
- "Negative D-dimer always rules out PE." Only when pretest probability is low; in high-probability patients you must image.
- "A clot in the leg means you have to find a clot in the lung." Most DVTs are not embolizing at the moment of diagnosis; absence of PE doesn't change DVT treatment.
- "Thrombolytics for everyone with PE." Reserved for hemodynamic instability or selected intermediate-high risk; major bleeding risk including 2-3% intracranial hemorrhage.
- "DOACs are always safer than warfarin." Antiphospholipid syndrome and mechanical valves still require warfarin.
- "Heparin and warfarin overlap is mandatory." Modern DOAC monotherapy (apixaban, rivaroxaban) skips parenteral lead-in for most PE patients.
Frequently asked questions
Where do pulmonary emboli come from?
More than 90% of PEs originate from deep vein thromboses in the proximal lower extremities — the popliteal, femoral, and iliac veins. A smaller fraction arises from pelvic veins (especially in postpartum and post-pelvic-surgery patients), upper-extremity catheters, and the right atrium in patients with atrial fibrillation. Rare non-thrombotic emboli include fat (long-bone fractures), amniotic fluid (during labor), air (central line misadventure), tumor (renal cell carcinoma extending up the IVC), and septic emboli (right-sided endocarditis from IV drug use). The leg-clot pathway is the dominant clinical entity and the target of all standard prophylaxis algorithms.
What does PE actually feel like?
The textbook triad — sudden dyspnea, pleuritic chest pain, and hemoptysis — appears together in only about 20% of cases. Far more common is acute unexplained shortness of breath, tachycardia (>100), and a low-grade fever; many patients also report a sense of impending doom. Massive PE presents with syncope, hypotension, and right heart strain. Small subsegmental PEs can be entirely silent and discovered only when CT is done for another reason. Pulse oximetry is unreliable — up to 25% of patients have normal SpO₂. This is why the Wells score and clinical gestalt matter so much: PE is the diagnosis you have to be willing to consider before you can find it.
How does the Wells score work?
The Wells score for PE assigns points for clinical features: signs of DVT (3), PE more likely than alternative diagnosis (3), heart rate >100 (1.5), immobilization or surgery in past 4 weeks (1.5), prior VTE (1.5), hemoptysis (1), cancer (1). Three-tier scoring: <2 low, 2-6 moderate, >6 high. The two-tier version splits at 4 into unlikely vs likely. Combined with a high-sensitivity D-dimer, low-probability patients with a negative D-dimer can have PE ruled out without imaging — saving hundreds of thousands of unnecessary CTs annually. The PERC rule (Pulmonary Embolism Rule-out Criteria) is an even stricter zero-step gate for very-low-probability outpatients.
Why is CTPA the gold standard?
CT pulmonary angiography is fast (seconds in the scanner), widely available, highly sensitive and specific (≥95% for segmental and larger PE), and offers alternative diagnoses on the same scan — pneumonia, pneumothorax, aortic dissection, malignancy. Limitations: contrast nephropathy in renal dysfunction, contrast allergy, radiation exposure (especially concerning in pregnancy and young women — breast dose is non-trivial), and reduced specificity for very small subsegmental emboli. V/Q scintigraphy is the alternative when contrast is contraindicated; it produces ventilation-perfusion mismatched defects but is harder to interpret and often non-diagnostic in patients with abnormal baseline lungs. Pulmonary angiography (direct catheterization) is the historical gold standard, now reserved for selected interventional cases.
What happens hemodynamically in massive PE?
Sudden mechanical obstruction of the pulmonary vasculature raises pulmonary vascular resistance. The thin-walled right ventricle, unaccustomed to high afterload, dilates and fails. Right ventricular pressure rises, septum bows leftward, left ventricular filling decreases, cardiac output drops, coronary perfusion of the RV falls, and a death spiral begins. Echocardiography shows RV dilation, hypokinesis, septal flattening, and the McConnell sign (apical sparing of RV free-wall hypokinesis — highly specific). Troponin and BNP elevate from RV strain and predict mortality. This is the population for thrombolytics: tPA (alteplase 100 mg over 2 hours) restores flow but at the cost of a 2-3% intracranial hemorrhage rate. Catheter-directed thrombolysis with lower drug doses is an alternative.
How is PE treated long-term?
Anticoagulation for at least 3 months, longer for unprovoked or recurrent events. First-line agents are direct oral anticoagulants — apixaban (10 mg BID for 7 days, then 5 mg BID), rivaroxaban (15 mg BID for 21 days, then 20 mg daily), dabigatran, and edoxaban (the latter two require initial parenteral lead-in). Warfarin (INR 2-3) is reserved for antiphospholipid syndrome and mechanical valves. LMWH (enoxaparin 1 mg/kg BID) is preferred in pregnancy and cancer. Reduced-dose apixaban (2.5 mg BID) or rivaroxaban (10 mg daily) extends therapy beyond 6 months in patients at intermediate bleeding risk. Stopping anticoagulation in unprovoked PE produces a 10% recurrence in the first year alone, so the decision to continue indefinitely is a shared-risk conversation.
What about IVC filters?
Inferior vena cava filters mechanically catch emboli ascending from the legs. Indications are now narrow: acute PE with an absolute contraindication to anticoagulation (recent CNS bleed, active major bleeding, planned imminent surgery) or recurrent PE despite therapeutic anticoagulation. The PREPIC trial showed early reduction in PE rate but increased DVT recurrence and no mortality benefit at 8 years. Retrievable filters must be removed within months — left in place they fracture, migrate, perforate the IVC, and become irretrievable nidus for thrombosis. Once anticoagulation can be safely resumed, the filter should come out.